Apple M4 Max (64 GB) vs Apple M1 Max (64 GB) for local AI
Pick two to four systems or devices to see which runs more models, faster, for the kind of work you do.
Too close to call
For general chat, the Apple M4 Max (64 GB) and Apple M1 Max (64 GB) come out about even. Look at price and the specific models you care about below.
Side by side
Swap sides All models on the M4 Max All models on the M1 Max
 Apple M4 Max (64 GB)
|
 Apple M1 Max (64 GB)
|
|
|---|---|---|
| Memory | 64.0 GB unified | 64.0 GB unified |
| Usable for models | 48.0 GB | 48.0 GB |
| Memory bandwidth | 546 GB/s | 400 GB/s |
| FP16 compute | 18.4 TFLOPS | 10.4 TFLOPS |
| Launch price | โ | โ |
| Released | 2024 | 2021 |
| LLMs that run | 521 of 622 | 521 of 622 |
| LLMs that run well | 504 | 485 |
| Image models that run | 61 of 61 | 61 of 61 |
| Video models that run | 42 of 43 | 42 of 43 |
| Decision models that run | 69 of 70 | 69 of 70 |
| Top-10 LLM average score | 100 | 100 |
| Top-10 LLM average speed | 89.4 tok/s | 71.8 tok/s |
| Best LLM | Qwen3.5 35B-A3B | Qwen3.5 35B-A3B |
Highlighted values are the best of the two. Model results assume 8K of context and the default macOS GPU memory limit on Macs.
LLMs
521 of 622 LLMs run on the Apple M4 Max (64 GB). 48.0 GB usable for models.
521 of 622 LLMs run on the Apple M1 Max (64 GB). 48.0 GB usable for models.
| Model | QualityM4 Max | SpeedM4 Max | RunsM4 Max | QualityM1 Max | SpeedM1 Max | RunsM1 Max |
|---|---|---|---|---|---|---|
| Qwen3.5 35B-A3B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 102 tok/s | Runs great | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 82.7 tok/s | Runs great |
| Qwen3.6 35B-A3B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 102 tok/s | Runs great | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 82.7 tok/s | Runs great |
| Sage-A1.5 35B-A3B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Flint (40B-A2.5B) | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 102 tok/s | Runs great | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 82.4 tok/s | Runs great |
| Ornith 1.5 35B-A3B | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| MiniCPM-V 4.7 35B-A3B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 102 tok/s | Runs great | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 82.7 tok/s | Runs great |
| Apodex 1.1 Mini (35B-A3B) | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Ornith 1.0 35B-A3B | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| GLM-4.7 Flash (30B-A3B) | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Raon-1 mini Preview (30B-A4B) | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 54.2 tok/s | Runs great | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 42.0 tok/s | Runs great |
| Nex N2 Mini (35B-A3B) | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 102 tok/s | Runs great | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 99/100 (quality, fit, speed and memory use) sets the ranking. | 82.7 tok/s | Runs great |
| Agents-A1 35B-A3B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Qwen AgentWorld 35B-A3B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Gemma 4 26B-A4B | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. | 76.3 tok/s | Runs great | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. | 60.4 tok/s | Runs great |
| Laguna XS 2.1 (33B-A3B) | 81Great quality (A) ยท 81 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 81Great quality (A) ยท 81 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Nemotron Cascade 2 30B-A3B | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. | 63.6 tok/s | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. | 49.8 tok/s | Runs great |
| Nanbeige4.3 33B | 94Excellent quality (A+) ยท 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. | 12.1 tok/s | Runs well | 92Excellent quality (A+) ยท 92 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 10.5 tok/s | Runs well |
| KAT-Coder V2.5 Dev (35B-A3B) | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| DiffusionGemma 26B-A4B IT | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. | 57.4 tok/s | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great |
| Nemotron 3 Nano 30B-A3B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. | 63.6 tok/s | Runs great | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. | 49.8 tok/s | Runs great |
| Qwen3 30B-A3B Instruct 2507 | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. | 66.5 tok/s | Runs great | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. | 52.3 tok/s | Runs great |
| Qwen3-VL 30B-A3B Thinking | 76Great quality (A) ยท 76 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. | 66.5 tok/s | Runs great | 76Great quality (A) ยท 76 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. | 52.3 tok/s | Runs great |
| Nemotron 3.5 Bolt 74B-A9B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 65.3 tok/s | Runs welltight fit | 77Great quality (A) ยท 77 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 50.7 tok/s | Runs welltight fit |
| Kolibri 1 (78B-A3.5B) | 76Great quality (A) ยท 76 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 127 tok/s | Runs welltight fit | 76Great quality (A) ยท 76 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 104 tok/s | Runs welltight fit |
| Qwen3 30B-A3B | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 66.5 tok/s | Runs great | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 52.3 tok/s | Runs great |
| Qwen3-VL 30B-A3B Instruct | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 66.5 tok/s | Runs great | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 52.3 tok/s | Runs great |
| Hunyuan Hy-Memory 30B-A3B | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| LFM2.5 24B-A2B | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 82.4 tok/s | Runs great | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 65.8 tok/s | Runs great |
| Nemotron 3.5 Lightning 30B-A3B | 72Great quality (A) ยท 72 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 72Great quality (A) ยท 72 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Kimi Linear 48B-A3B | 71Great quality (A) ยท 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 102 tok/s | Runs great | 71Great quality (A) ยท 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 82.7 tok/s | Runs great |
| Qwen3.8 27B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 20.8 tok/s | Runs well | 94Excellent quality (A+) ยท 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs well |
| Qwen3.6 27B | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 20.8 tok/s | Runs well | 93Excellent quality (A+) ยท 93 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs well |
| Qwen3-Next 80B-A3B Thinking | 72Great quality (A) ยท 72 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 138 tok/s | Runs welltight fit | 72Great quality (A) ยท 72 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 115 tok/s | Runs welltight fit |
| Sarvam 30B | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 83.5 tok/s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 66.9 tok/s | Runs great |
| Qwen3-Omni 30B-A3B Instruct | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 84.4 tok/s | Runs great | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 67.3 tok/s | Runs great |
| Gemma 4 31B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 18.5 tok/s | Runs well | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 11.2 tok/s | Runs well |
| Qwen3.5 27B | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 20.8 tok/s | Runs well | 92Excellent quality (A+) ยท 92 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs well |
| North Mini Code 1.0 (30B-A3B) | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Qwen3 Next 80B-A3B Instruct | 70Great quality (A) ยท 70 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 138 tok/s | Runs welltight fit | 70Great quality (A) ยท 70 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 115 tok/s | Runs welltight fit |
| Muse Glimmer 30B | 87Excellent quality (A+) ยท 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 19.4 tok/s | Runs well | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 11.7 tok/s | Runs well |
| Qwen3 30B-A3B Thinking 2507 | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 66.5 tok/s | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 52.3 tok/s | Runs great |
| Qwen3 Coder 30B-A3B | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 66.5 tok/s | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 52.3 tok/s | Runs great |
| TMax 27B | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 21.1 tok/s | Runs well | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 10.7 tok/s | Runs well |
| Trinity Mini (26B-A3B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Fara 1.5 27B | 83Great quality (A) ยท 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 21.1 tok/s | Runs well | 87Excellent quality (A+) ยท 87 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 10.7 tok/s | Runs well |
| UI-Mate 27B | 81Great quality (A) ยท 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 21.1 tok/s | Runs well | 81Great quality (A) ยท 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 15.5 tok/s | Runs well |
| Gemma 4 12B | 80Great quality (A) ยท 80 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 39.9 tok/s | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 35.4 tok/s | Runs great |
| Axion 1.0 35B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 61Good quality (B) ยท 61 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Seed-OSS 36B Instruct | 76Great quality (A) ยท 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 16.0 tok/s | Runs well | 76Great quality (A) ยท 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 11.7 tok/s | Runs well |
| ERNIE 4.5 21B-A3B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 64Good quality (B) ยท 64 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| EXAONE 4.5 33B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 16.8 tok/s | Runs well | 77Great quality (A) ยท 77 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 12.3 tok/s | Runs well |
| Llama 3.3 Nemotron Super 49B v1.5 | 76Great quality (A) ยท 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 11.6 tok/s | Runs well | 67Good quality (B) ยท 67 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 11.0 tok/s | Runs well |
| Granite 4.0 H Small (32B-A9B) | 61Good quality (B) ยท 61 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 38.3 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great |
| Hermes 4.3 36B | 75Great quality (A) ยท 75 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 16.0 tok/s | Runs well | 75Great quality (A) ยท 75 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 11.7 tok/s | Runs well |
| Hunyuan A13B (80B-A13B) | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 48.4 tok/s | Runs welltight fit | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 37.0 tok/s | Runs welltight fit |
| K2 Horizon 7B | 79Great quality (A) ยท 79 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 44.5 tok/s | Runs great | 77Great quality (A) ยท 77 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 38.8 tok/s | Runs great |
| Qwen3 Coder Next (80B-A3B) | 72Great quality (A) ยท 72 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 138 tok/s | Runs welltight fit | 72Great quality (A) ยท 72 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 115 tok/s | Runs welltight fit |
| NorthCUA Mini | 58Good quality (B) ยท 58 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 90.1 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 72.3 tok/s | Runs great |
| Qwen3-VL 32B Instruct | 74Great quality (A) ยท 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 17.3 tok/s | Runs well | 74Great quality (A) ยท 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 12.7 tok/s | Runs well |
| InternVL3.5 38B | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 15.0 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 11.0 tok/s | Runs well |
| SenseNova Loop 9B Preview | 77Great quality (A) ยท 77 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 41.5 tok/s | Runs great | 74Great quality (A) ยท 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great |
| Mellum2.1 Thinking (12B-A2.5B) | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 64.9 tok/s | Runs great |
| Ornith 1.5 9B | 76Great quality (A) ยท 76 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 41.5 tok/s | Runs great | 73Great quality (A) ยท 73 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great |
| Qwen3 32B | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| ERNIE 4.5 21B-A3B Thinking | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Phi-3.5 MoE (42B-A6.6B) | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 49.8 tok/s | Runs great | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 38.3 tok/s | Runs great |
| Qwen3.5 9B | 75Great quality (A) ยท 75 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 41.5 tok/s | Runs great | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great |
| Granite 4.2 30B | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 19.7 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 14.4 tok/s | Runs well |
| LFM2 24B-A2B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 86.0 tok/s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 69.0 tok/s | Runs great |
| Ornith 1.0 9B | 74Great quality (A) ยท 74 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 41.5 tok/s | Runs great | 71Great quality (A) ยท 71 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great |
| gpt-oss 20B | 70Great quality (A) ยท 70 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 109 tok/s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 88.3 tok/s | Runs great |
| Ling 2.6 Flash (104B-A7.4B) | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 91.5 tok/s | Runs welltight fit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 72.6 tok/s | Runs welltight fit |
| Solar Open 100B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 63.4 tok/s | Runs welltight fit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 49.0 tok/s | Runs welltight fit |
| Maple Preview (20B-A1B) | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 113 tok/s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 93.2 tok/s | Runs great |
| Devstral Small 2 (24B) | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 24.0 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well |
| TMax 9B | 73Great quality (A) ยท 73 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great | 71Great quality (A) ยท 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 39.0 tok/s | Runs great |
| GLM-4 32B 0414 | 70Great quality (A) ยท 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 17.7 tok/s | Runs well | 70Great quality (A) ยท 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 13.0 tok/s | Runs well |
| InternVL3 38B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 15.0 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 11.0 tok/s | Runs well |
| Ling Mini 2.0 (16B-A1.4B) | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 68.0 tok/s | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 54.1 tok/s | Runs great |
| Granite 4.1 30B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 20.0 tok/s | Runs well | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 14.6 tok/s | Runs well |
| Ling Flash 2.0 (100B-A6B) | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 105 tok/s | Runs welltight fit | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 84.0 tok/s | Runs welltight fit |
| Apriel 1.5 15B Thinker | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 38.8 tok/s | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 28.5 tok/s | Runs well |
| Nemotron 3.5 Bolt E4B | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 65.3 tok/s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 47.8 tok/s | Runs great |
| Groot 19B A3B 261023 | 57Good quality (B) ยท 57 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 57Good quality (B) ยท 57 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Olmo 3.1 32B Instruct | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 13.1 tok/s | Runs well |
| EXAONE 4.0 32B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 18.0 tok/s | Runs well | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 13.2 tok/s | Runs well |
| Qwen2.5 32B Instruct | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Baichuan M2 32B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Sarvam 105B | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 73.1 tok/s | Runs welltight fit | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 57.1 tok/s | Runs welltight fit |
| Kimi-VL A3B Thinking 2506 | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Gemma 3 27B IT | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 21.1 tok/s | Runs well | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 15.4 tok/s | Runs well |
| Nous Hermes 2 Mixtral 8x7B DPO | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 38.9 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 29.5 tok/s | Runs well |
| Phi-4 Reasoning Vision 15B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 38.2 tok/s | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 28.0 tok/s | Runs well |
| Qwen2.5-VL 32B Instruct | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 17.3 tok/s | Runs well | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 12.6 tok/s | Runs well |
| Gemma 4 E4B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 65.3 tok/s | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 47.8 tok/s | Runs great |
| Ling 3.0 Tiny (8B-A1.3B) | 60Good quality (B) ยท 60 of 100: how capable the model is at F16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 71.6 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at F16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 57.2 tok/s | Runs great |
| Granite 4.2 8B | 67Good quality (B) ยท 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 45.6 tok/s | Runs great | 65Good quality (B) ยท 65 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 39.7 tok/s | Runs great |
| Jamba 1.5 Mini | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 41.4 tok/s | Runs great | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 31.5 tok/s | Runs great |
| Devstral Small 1.1 (24B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 24.5 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well |
| Magistral Small 2507 | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 24.5 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well |
| MedGemma 27B Text IT | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 21.4 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 15.7 tok/s | Runs well |
| Mixtral 8x7B Instruct | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 38.9 tok/s | Runs great | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 29.5 tok/s | Runs well |
| QwQ 32B | 74Great quality (A) ยท 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 74Great quality (A) ยท 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Qwen3 14B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 39.0 tok/s | Runs great | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| MiMo V2.6 Distill Qwen 9B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 75Great quality (A) ยท 75 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 37.1 tok/s | Runs great |
| Qwen2 57B-A14B Instruct | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 36.2 tok/s | Runs great | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 27.4 tok/s | Runs well |
| Nanbeige4.2 3B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 70.5 tok/s | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 51.6 tok/s | Runs great |
| Mistral Small 3.2 24B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 24.0 tok/s | Runs well | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well |
| Magistral Small 2506 | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 24.5 tok/s | Runs well | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well |
| Terminal RSI 27B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 20.8 tok/s | Runs well | 64Good quality (B) ยท 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 15.2 tok/s | Runs well |
| OpenReasoning Nemotron 32B | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Skywork OR1 32B | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Hermes 4 14B | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| Gemma 4 E2B | 62Good quality (B) ยท 62 of 100: how capable the model is at F16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 58.9 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at F16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 43.1 tok/s | Runs great |
| Nemotron Labs Dory 1.0 4B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 73.8 tok/s | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 54.1 tok/s | Runs great |
| Hemmingway 1.5 | 63Good quality (B) ยท 63 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 21.5 tok/s | Runs well | 63Good quality (B) ยท 63 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 15.7 tok/s | Runs well |
| Falcon-H1 34B Instruct | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 17.2 tok/s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 12.6 tok/s | Runs well |
| Hermes 4 70B | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Llama 3.3 70B Instruct | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Mistral Small 3.1 24B | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 24.0 tok/s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well |
| Phi-4 (14B) | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 39.3 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 28.8 tok/s | Runs well |
| Olmo 3.1 32B Think | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 13.1 tok/s | Runs well |
| InternVL3 14B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 38.2 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 28.0 tok/s | Runs well |
| EXAONE Deep 32B | 70Great quality (A) ยท 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 18.0 tok/s | Runs well | 70Great quality (A) ยท 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 13.2 tok/s | Runs well |
| Ministral 3 14B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 41.4 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 30.3 tok/s | Runs great |
| Qwen2.5 72B Instruct | 61Good quality (B) ยท 61 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 10.3 tok/s | Runs welltight fit | 61Good quality (B) ยท 61 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 7.5 tok/s | Slowtight fit |
| Qwen2.5-VL 72B Instruct | 61Good quality (B) ยท 61 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 10.2 tok/s | Runs welltight fit | 61Good quality (B) ยท 61 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 7.4 tok/s | Slowtight fit |
| Spark-X2.6 4B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 71.5 tok/s | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 52.4 tok/s | Runs great |
| DeepSeek R1 Distill Llama 70B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 60Good quality (B) ยท 60 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Tรผlu 3 70B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 60Good quality (B) ยท 60 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Qwen3-VL 8B Instruct | 61Good quality (B) ยท 61 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 45.7 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 39.8 tok/s | Runs great |
| Mistral Small 3 24B (2501) | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 24.5 tok/s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well |
| Command R (08-2024) 32B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 13.1 tok/s | Runs well |
| LFM2.5 8B-A1B | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 64.7 tok/s | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 51.3 tok/s | Runs great |
| InternVL3.5 8B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 46.9 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 40.9 tok/s | Runs great |
| Qwen2.5 14B Instruct | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 39.3 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 28.8 tok/s | Runs well |
| Ring Flash 2.0 (100B-A6B) | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 105 tok/s | Runs welltight fit | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 84.0 tok/s | Runs welltight fit |
| GLM-Z1 32B (0414) | 70Great quality (A) ยท 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 17.7 tok/s | Runs well | 70Great quality (A) ยท 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 13.0 tok/s | Runs well |
| DeepSeek R1 Distill Qwen 32B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| DeepSeek Coder V2 Lite (16B) | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 45.3 tok/s | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 66.9 tok/s | Runs great |
| Gemma 2 27B IT | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 21.2 tok/s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 15.5 tok/s | Runs well |
| Llama 3.1 70B Instruct | 58Good quality (B) ยท 58 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 58Good quality (B) ยท 58 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Kimi-VL A3B Instruct | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Aya Vision 32B | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 17.4 tok/s | Runs well | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 12.8 tok/s | Runs well |
| DeepSeek V2 Lite Chat | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 45.3 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 66.9 tok/s | Runs great |
| Granite 4.1 8B | 59Good quality (B) ยท 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 45.6 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 39.7 tok/s | Runs great |
| Voxtral Small 24B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 23.8 tok/s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 17.4 tok/s | Runs well |
| Dolphin Mistral 24B Venice Edition | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 24.0 tok/s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well |
| Aya Expanse 32B | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 13.1 tok/s | Runs well |
| Phi-4 Reasoning Plus (14B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 39.4 tok/s | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 28.8 tok/s | Runs well |
| GLM-4.1V 9B Thinking | 59Good quality (B) ยท 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 38.9 tok/s | Runs great | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 41.1 tok/s | Runs great |
| MiniCPM-V 4.5 8B | 59Good quality (B) ยท 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 46.0 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 40.1 tok/s | Runs great |
| OLMo 2 32B Instruct | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 13.1 tok/s | Runs well |
| Command R 35B (v01) | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 16.5 tok/s | Runs well | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 12.1 tok/s | Runs well |
| Qwen2.5 14B Instruct 1M | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| Qwen3 8B | 59Good quality (B) ยท 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 48.9 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great |
| OpenHands LM 32B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Gemma 3 12B IT | 58Good quality (B) ยท 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 34.7 tok/s | Runs great |
| Qwen3.5 4B | 62Good quality (B) ยท 62 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 63.0 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 46.2 tok/s | Runs great |
| Nemotron Nano 9B v2 | 59Good quality (B) ยท 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 45.0 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 39.3 tok/s | Runs great |
| Hermes 3 (Llama 3.1 70B) | 57Good quality (B) ยท 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 57Good quality (B) ยท 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| QVQ 72B Preview | 57Good quality (B) ยท 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 10.2 tok/s | Runs welltight fit | 57Good quality (B) ยท 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 7.4 tok/s | Slowtight fit |
| Magistral Small 1.2 (24B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 24.0 tok/s | Runs well | 68Good quality (B) ยท 68 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well |
| UI-Mate 9B | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 67Good quality (B) ยท 67 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 37.1 tok/s | Runs great |
| UI-Venus 2 9B | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 67Good quality (B) ยท 67 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 37.1 tok/s | Runs great |
| Dolphin 3.0 R1 Mistral 24B | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 24.5 tok/s | Runs well | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well |
| Phi-4 Reasoning (14B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 39.4 tok/s | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 28.8 tok/s | Runs well |
| Solar Pro Preview 22B | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 26.1 tok/s | Runs well | 57Good quality (B) ยท 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 19.1 tok/s | Runs well |
| TMax 4B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 69.8 tok/s | Runs great | 61Good quality (B) ยท 61 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 51.1 tok/s | Runs great |
| Kimi Dev 72B | 67Good quality (B) ยท 67 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 10.3 tok/s | Runs welltight fit | 67Good quality (B) ยท 67 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 7.5 tok/s | Slowtight fit |
| MiMo-VL 7B RL | 57Good quality (B) ยท 57 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 48.2 tok/s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 42.0 tok/s | Runs great |
| MiniCPM5 2B | 60Good quality (B) ยท 60 of 100: how capable the model is at F16, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 53.7 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 85.4 tok/s | Runs great |
| Llama 3.1 Nemotron 70B Instruct | 55Good quality (B) ยท 55 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 55Good quality (B) ยท 55 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Apertus v1.5 70B | 55Good quality (B) ยท 55 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 10.4 tok/s | Runs welltight fit | 55Good quality (B) ยท 55 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 7.6 tok/s | Slowtight fit |
| Agents-A1 4B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 64.7 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 47.4 tok/s | Runs great |
| InternVL3 78B | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 12.1 tok/s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 7.0 tok/s | Slowtight fit |
| Qwen2.5 Coder 32B Instruct | 63Good quality (B) ยท 63 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 63Good quality (B) ยท 63 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Ministral 3 8B | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 44.9 tok/s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 39.2 tok/s | Runs great |
| Moonlight 16B-A3B Instruct | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 71.4 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 tok/s | Runs great |
| Aya 23 35B | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 16.5 tok/s | Runs well | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 12.1 tok/s | Runs well |
| QwQ 32B Preview | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 tok/s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 tok/s | Runs well |
| Qwen2 72B Instruct | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 10.3 tok/s | Runs welltight fit | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 7.5 tok/s | Slowtight fit |
| Qwen2-VL 72B Instruct | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 10.2 tok/s | Runs welltight fit | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 7.4 tok/s | Slowtight fit |
| MiMo 7B RL | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 51.1 tok/s | Runs great | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 37.5 tok/s | Runs great |
| Fara 1.5 4B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 64.7 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 47.4 tok/s | Runs great |
| InternVL3 8B | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 50.4 tok/s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 44.0 tok/s | Runs great |
| Mage-VL 4.7B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 62.0 tok/s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great |
| AceReason Nemotron 14B | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| OpenReasoning Nemotron 14B | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| Qwen2.5 Coder 14B Instruct | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| Apriel Nemotron 15B Thinker | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 38.6 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 28.2 tok/s | Runs well |
| Ministral 3 14B Reasoning | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 41.4 tok/s | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 30.3 tok/s | Runs great |
| Llama 3.1 Nemotron Nano 8B v1 | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Falcon 3 10B Instruct | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 38.8 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 41.0 tok/s | Runs great |
| Mistral Small 22B (2409) | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 25.9 tok/s | Runs well | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 19.0 tok/s | Runs well |
| Hunyuan 7B Instruct | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 39.2 tok/s | Runs great | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great |
| Devstral Small 2505 | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 24.5 tok/s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 17.9 tok/s | Runs well |
| MiniCPM4 8B | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 48.9 tok/s | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great |
| SenseNova 6.8 Code2Video | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 21.1 tok/s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 15.5 tok/s | Runs well |
| Reka Flash 3.1 (21B) | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 27.6 tok/s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 20.2 tok/s | Runs well |
| Mistral Nemo 12B Instruct | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 34.7 tok/s | Runs great |
| DeepCoder 14B Preview | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| Apertus 70B Instruct | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| Qwen3-VL 4B Instruct | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 66.2 tok/s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Runs great |
| GLM-4 9B 0414 | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 45.0 tok/s | Runs great |
| NorthCUA Nano | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 44.9 tok/s | Runs great |
| Olmo 3 7B Instruct | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great |
| Qwen3 4B Instruct 2507 | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 73.1 tok/s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 53.5 tok/s | Runs great |
| LFM3 1.7B | 56Good quality (B) ยท 56 of 100: how capable the model is at F16, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 78.3 tok/s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at F16, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 57.3 tok/s | Runs great |
| MiniCPM-o 2.6 8B | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 46.2 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 40.3 tok/s | Runs great |
| Gemma 2 9B IT | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 43.3 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 37.8 tok/s | Runs great |
| Yi 1.5 34B Chat | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 16.8 tok/s | Runs well | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 12.3 tok/s | Runs well |
| Qwen2.5-VL 7B Instruct | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 48.3 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 42.1 tok/s | Runs great |
| Granite 4.0 H Tiny (7B-A1B) | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 85.0 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 69.2 tok/s | Runs great |
| Llama 3 70B Instruct | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 10.6 tok/s | Runs welltight fit | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 7.7 tok/s | Slowtight fit |
| DeepSeek R1 Distill Qwen 14B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| DeepSeek R1 0528 Qwen3 8B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 48.9 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great |
| Pixtral 12B (2409) | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 34.1 tok/s | Runs great |
| Qwen2.5 7B Instruct | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| JanusCoder 14B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 28.6 tok/s | Runs well |
| Codestral 22B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 25.9 tok/s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 19.0 tok/s | Runs well |
| InternVL2.5 8B | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 49.6 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 43.2 tok/s | Runs great |
| Foundation-Sec 1.1 8B Instruct | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| InternLM3 8B Instruct | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 45.5 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 39.7 tok/s | Runs great |
| InternLM2.5 20B Chat | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 29.1 tok/s | Runs well | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 21.3 tok/s | Runs well |
| Granite 4.2 3B | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 80.3 tok/s | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 58.8 tok/s | Runs great |
| GLiDE 9B | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 50.4 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 44.0 tok/s | Runs great |
| Qwen2.5 7B Instruct 1M | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| Falcon-H1 7B Instruct | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 52.8 tok/s | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 38.7 tok/s | Runs great |
| GLM-4 9B Chat 1M | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 42.2 tok/s | Runs great | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 44.6 tok/s | Runs great |
| Llama 3.2 90B Vision Instruct | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 10.7 tok/s | Runs welltight fit | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 7.8 tok/s | Slowtight fit |
| GLM-4 9B Chat (2024) | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 45.0 tok/s | Runs great |
| LFM2 8B-A1B | 38Basic quality (D) ยท 38 of 100: how capable the model is at F16, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 64.7 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at F16, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 51.3 tok/s | Runs great |
| Qwen3 4B | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 73.1 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 53.5 tok/s | Runs great |
| Hermes 3 (Llama 3.1 8B) | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Llama 3.1 8B Instruct | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Mage-VL-Cowork 4.7B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 62.0 tok/s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great |
| Falcon 3 7B Instruct | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 39.4 tok/s | Runs great | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 39.3 tok/s | Runs great |
| MiniCPM-V 2.6 8B | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.4 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 43.1 tok/s | Runs great |
| Gemma 3n E4B | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 73.5 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 53.8 tok/s | Runs great |
| GLM-Z1 9B (0414) | 57Good quality (B) ยท 57 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 55Good quality (B) ยท 55 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 45.0 tok/s | Runs great |
| Seed-Coder 8B Reasoning | 57Good quality (B) ยท 57 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 42.3 tok/s | Runs great |
| Nemotron 3 Nano 4B | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 74.0 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 54.2 tok/s | Runs great |
| Phi-3 Medium 128K Instruct | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 41.3 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 30.3 tok/s | Runs great |
| Apertus v1.5 8B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 45.0 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 39.2 tok/s | Runs great |
| OLMo 2 13B Instruct | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 42.1 tok/s | Runs great | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 30.8 tok/s | Runs great |
| LFM3-VL 2B | 50Fair quality (C) ยท 50 of 100: how capable the model is at F16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 70.5 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at F16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 51.7 tok/s | Runs great |
| Qwen1.5 32B Chat | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 17.8 tok/s | Runs well | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 13.0 tok/s | Runs well |
| Granite 3.2 8B Instruct | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.0 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 42.8 tok/s | Runs great |
| Granite 3.3 8B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.0 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 42.8 tok/s | Runs great |
| LLM4Decompile 22B V2 | 55Good quality (B) ยท 55 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 25.9 tok/s | Runs well | 55Good quality (B) ยท 55 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 19.0 tok/s | Runs well |
| Ministral 8B (2410) | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great |
| DeepHermes 3 Llama 3 8B Preview | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Dolphin 3.0 Llama 3.1 8B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Tรผlu 3 8B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Phi-3 Medium (14B) | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 41.3 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 30.3 tok/s | Runs great |
| LLaVA 1.6 34B | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 16.6 tok/s | Runs well | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 12.2 tok/s | Runs well |
| EXAONE 3.5 7.8B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 51.2 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great |
| Phi-3 Small 8K Instruct | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 39.8 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 39.7 tok/s | Runs great |
| GLM-4.6V | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 53.8 tok/s | Runs wellover GPU memory limit | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 41.7 tok/s | Runs wellover GPU memory limit |
| Falcon-H1R 7B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 38.7 tok/s | Runs great | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 38.7 tok/s | Runs great |
| Qwen2.5-Omni 7B | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 44.4 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 39.4 tok/s | Runs great |
| Olmo 3 7B Think | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great |
| Seed-Coder 8B Instruct | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 42.3 tok/s | Runs great |
| Qwen1.5 MoE A2.7B Chat | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 41.1 tok/s | Runs great | 24Basic quality (D) ยท 24 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 61.2 tok/s | Runs great |
| EXAONE Deep 7.8B | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 51.2 tok/s | Runs great | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 37.5 tok/s | Runs great |
| Granite 3.1 8B Instruct | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 49.0 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 42.8 tok/s | Runs great |
| InternVL2 8B | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 49.6 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 43.2 tok/s | Runs great |
| Dolphin 2.9.4 Llama 3.1 8B | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Hermes 2 Pro Llama 3 8B | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| OpenChat 3.6 8B | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Mistral Small 4 (119B-A6B) | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 89.9 tok/s | Runs wellover GPU memory limit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 72.2 tok/s | Runs wellover GPU memory limit |
| Aya Vision 8B | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 46.4 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Phi-4 Multimodal Instruct | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 52.7 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 52.7 tok/s | Runs great |
| Qwen2-VL 7B Instruct | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 48.3 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 42.1 tok/s | Runs great |
| GLM-4.5V | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 53.8 tok/s | Runs wellover GPU memory limit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 41.7 tok/s | Runs wellover GPU memory limit |
| Command R7B | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Jamba Reasoning 3B | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 42.3 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 67.3 tok/s | Runs great |
| LFM2.5-VL 3B | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 43.4 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 69.0 tok/s | Runs great |
| Qwen2 7B Instruct | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| Qwen3 4B Thinking 2507 | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 73.1 tok/s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 53.5 tok/s | Runs great |
| Granite 4.1 3B | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 86.4 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 63.3 tok/s | Runs great |
| Qwen2.5 Coder 7B Instruct | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| Mistral-NeMo-Minitron 8B Instruct | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 47.6 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 41.5 tok/s | Runs great |
| North Micro Vision | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 54.6 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 40.0 tok/s | Runs great |
| Llama 3.2 11B Vision Instruct | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 39.6 tok/s | Runs great |
| Llama 3.1 Nemotron Nano 4B v1.1 | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 65.1 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 47.7 tok/s | Runs great |
| Gemma 3 4B IT | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 68.3 tok/s | Runs great | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 50.1 tok/s | Runs great |
| LFM2.5 2.6B | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 50.2 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 79.7 tok/s | Runs great |
| GLM-4.5 Air (106B-A12B) | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 53.8 tok/s | Runs wellover GPU memory limit | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 41.7 tok/s | Runs wellover GPU memory limit |
| Llama 3 8B Instruct | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Yi 34B Chat | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 16.8 tok/s | Runs well | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 12.3 tok/s | Runs well |
| InternLM2 Chat 20B | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 29.1 tok/s | Runs well | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 21.3 tok/s | Runs well |
| OpenReasoning Nemotron 7B | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| Phi-4 Mini (3.8B) | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 76.5 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 56.1 tok/s | Runs great |
| Ministral 3 3B | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 76.3 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 55.9 tok/s | Runs great |
| SmolLM3 3B | 42Fair quality (C) ยท 42 of 100: how capable the model is at F16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 44.0 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 69.9 tok/s | Runs great |
| MedGemma 4B IT | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 68.3 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 50.1 tok/s | Runs great |
| Command R7B Arabic | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| CogVLM2 Llama 3 Chat 19B | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 29.6 tok/s | Runs well | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 21.7 tok/s | Runs well |
| Qwen1.5 14B Chat | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 40.7 tok/s | Runs great | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 38.6 tok/s | Runs great |
| Foundation-Sec 8B Reasoning | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Jamba2 3B | 42Fair quality (C) ยท 42 of 100: how capable the model is at F16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 71.0 tok/s | Runs great |
| Hunyuan 4B Instruct | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 69.6 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 51.0 tok/s | Runs great |
| Nous Hermes 2 SOLAR 10.7B | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 44.4 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 39.4 tok/s | Runs great |
| SOLAR 10.7B Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 44.4 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 39.4 tok/s | Runs great |
| Apertus 8B Instruct | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 49.7 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 43.4 tok/s | Runs great |
| Aya Expanse 8B | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Phi-3.5 Mini Instruct | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 76.9 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 56.3 tok/s | Runs great |
| Gemma 3n E2B | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 67.7 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 49.6 tok/s | Runs great |
| Nemotron 3.5 Super VL 120B-A12B | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 53.8 tok/s | Slowfar over GPU memory limit | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 41.7 tok/s | Slowfar over GPU memory limit |
| InternLM2.5 7B Chat | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 51.7 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 45.1 tok/s | Runs great |
| Qwen2.5-Math 72B Instruct | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 10.3 tok/s | Runs welltight fit | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 7.5 tok/s | Slowtight fit |
| Granite 3.0 8B Instruct | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 49.0 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 42.8 tok/s | Runs great |
| Mistral 7B Instruct v0.3 | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Zamba2 7B Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 39.9 tok/s | Runs great | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 39.9 tok/s | Runs great |
| Leanstral 119B-A6B | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 85.1 tok/s | Runs wellover GPU memory limit | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 68.1 tok/s | Runs wellover GPU memory limit |
| Ornith 1.5 122B-A10B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 62.2 tok/s | Slowfar over GPU memory limit | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Slowfar over GPU memory limit |
| DeepSeek R1 Distill Llama 8B | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| OLMo 2 7B Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great |
| DeepSeek LLM 67B Chat | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 11.1 tok/s | Runs well | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 10.3 tok/s | Runs well |
| Qwen3.5 122B-A10B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 62.2 tok/s | Slowfar over GPU memory limit | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Slowfar over GPU memory limit |
| Falcon3 Mamba 7B Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.4 tok/s | Runs great | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.4 tok/s | Runs great |
| Fara 7B | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 48.3 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 42.1 tok/s | Runs great |
| Phi-3.5 Vision Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 70.8 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 51.9 tok/s | Runs great |
| AFM 4.5B | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 63.6 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 46.6 tok/s | Runs great |
| Dolphin 2.9 Llama 3 8B | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Llama 4 Scout 17B-16E | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.4 tok/s | Runs wellover GPU memory limit | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 30.8 tok/s | Runs wellover GPU memory limit |
| DeepSeek MoE 16B Chat | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 75.0 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 59.5 tok/s | Runs great |
| Qianfan-OCR 4B | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 62.0 tok/s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great |
| Hermes 2 Pro Mistral 7B | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Starling LM 7B Beta | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| EXAONE 3.0 7.8B Instruct | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 51.2 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great |
| Qwen3.5 2B | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 59.7 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 43.7 tok/s | Runs great |
| Granite 4.0 H Micro | 38Basic quality (D) ยท 38 of 100: how capable the model is at F16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 42.4 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 67.5 tok/s | Runs great |
| Granite 4.0 Micro (3B) | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 86.4 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 63.3 tok/s | Runs great |
| OpenChat 3.5 (0106) | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Groot 65B A6B 261023 | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 89.9 tok/s | Runs wellover GPU memory limit | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 72.2 tok/s | Runs wellover GPU memory limit |
| TMax 2B | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 72.0 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 52.8 tok/s | Runs great |
| Yi 1.5 9B Chat | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 39.6 tok/s | Runs great |
| Phi-3 Mini 128K Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 76.9 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 56.3 tok/s | Runs great |
| Nemotron 3 Super 120B-A12B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 53.8 tok/s | Slowfar over GPU memory limit | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 41.7 tok/s | Slowfar over GPU memory limit |
| MiniCPM-V 4.7 1B | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 87.4 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 64.0 tok/s | Runs great |
| Mistral 7B Instruct v0.2 | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Nous Hermes 2 Mistral 7B DPO | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| OpenChat 3.5 (1210) | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| DeepSeek Coder 33B Instruct | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 17.3 tok/s | Runs well | 41Fair quality (C) ยท 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 12.7 tok/s | Runs well |
| Chandra OCR 2 | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 55.4 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great |
| CodeGeeX4 All 9B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 45.0 tok/s | Runs great |
| Falcon-H1 3B Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 43.0 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 68.3 tok/s | Runs great |
| MiniCPM3 4B | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 72.2 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 52.9 tok/s | Runs great |
| Llama 3.2 3B Instruct | 35Basic quality (D) ยท 35 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 42.2 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 67.1 tok/s | Runs great |
| Llama 2 70B Chat | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 10.8 tok/s | Runs well | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 10.1 tok/s | Runs well |
| S5 nano | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 63.0 tok/s | Runs great | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 46.2 tok/s | Runs great |
| Voxtral Mini 3B | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 62.8 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 46.0 tok/s | Runs great |
| LightOnOCR-3 4B | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 64.7 tok/s | Runs great | 48Fair quality (C) ยท 48 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 47.4 tok/s | Runs great |
| RecurrentGemma 9B IT | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 41.6 tok/s | Runs great | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 43.9 tok/s | Runs great |
| Phi-3 Vision 128K Instruct | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 70.8 tok/s | Runs great | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 51.9 tok/s | Runs great |
| Phi-3 Mini 4K Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 76.9 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 56.3 tok/s | Runs great |
| Aya 23 8B | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 49.9 tok/s | Runs great | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| DeepSeek R1 Distill Qwen 7B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| Qwen3-VL 2B Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 63.6 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 46.6 tok/s | Runs great |
| OLMoE 1B-7B Instruct | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 71.6 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 57.2 tok/s | Runs great |
| OpenHermes 2.5 Mistral 7B | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Starling LM 7B Alpha | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| DeepSeek Prover V2 7B | 40Fair quality (C) ยท 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 42.5 tok/s | Runs great | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 42.5 tok/s | Runs great |
| Qwen2.5 3B Instruct | 34Basic quality (D) ยท 34 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 69.7 tok/s | Runs great |
| WeVisOne-VL 4B | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 60.8 tok/s | Runs great | 45Fair quality (C) ยท 45 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 44.6 tok/s | Runs great |
| Unlimited OCR 2 (3B-A0.6B) | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 114 tok/s | Runs great | 44Fair quality (C) ยท 44 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 95.9 tok/s | Runs great |
| Phi-4 Mini Reasoning (3.8B) | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 76.5 tok/s | Runs great | 46Fair quality (C) ยท 46 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 56.1 tok/s | Runs great |
| Hermes 3 Llama 3.2 3B | 32Basic quality (D) ยท 32 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 42.2 tok/s | Runs great | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 67.1 tok/s | Runs great |
| LFM2 2.6B | 34Basic quality (D) ยท 34 of 100: how capable the model is at F16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 52.7 tok/s | Runs great | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 83.8 tok/s | Runs great |
| Mistral 7B Instruct v0.1 | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 29Basic quality (D) ยท 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Ling 3.0 Flash (127B-A5B) | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 99.9 tok/s | Slowfar over GPU memory limit | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 81.2 tok/s | Slowfar over GPU memory limit |
| Raon 1 nano | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 83.2 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 61.0 tok/s | Runs great |
| Gemma 1.1 7B IT | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 46.9 tok/s | Runs great | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 40.9 tok/s | Runs great |
| Orca 2 13B | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 44.3 tok/s | Runs great | 23Basic quality (D) ยท 23 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 42.0 tok/s | Runs great |
| Qwen2.5-VL 3B Instruct | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 78.3 tok/s | Runs great | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 57.4 tok/s | Runs great |
| MiniCPM5 1B | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 125 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 91.9 tok/s | Runs great |
| Code Llama 70B Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 10.8 tok/s | Runs well | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 10.1 tok/s | Runs well |
| Falcon Mamba 7B Instruct | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 40.4 tok/s | Runs great | 29Basic quality (D) ยท 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 40.4 tok/s | Runs great |
| Granite Code 34B Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 17.1 tok/s | Runs well | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 12.5 tok/s | Runs well |
| Falcon 3 3B Instruct | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 41.9 tok/s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 66.6 tok/s | Runs great |
| Neural Chat 7B v3.3 | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Baichuan 2 13B Chat | 24Basic quality (D) ยท 24 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 41.5 tok/s | Runs great | 21Basic quality (D) ยท 21 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 39.3 tok/s | Runs great |
| Gemma 7B IT | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 46.9 tok/s | Runs great | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 40.9 tok/s | Runs great |
| Code Llama 34B Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 17.1 tok/s | Runs well | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 12.5 tok/s | Runs well |
| Qwen1.5 7B Chat | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 51.9 tok/s | Runs great | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 45.2 tok/s | Runs great |
| Yi Coder 9B Chat | 39Basic quality (D) ยท 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 39.6 tok/s | Runs great |
| StableLM 2 12B Chat | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 39.3 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 34.8 tok/s | Runs great |
| Llama 2 13B Chat | 24Basic quality (D) ยท 24 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 44.3 tok/s | Runs great | 21Basic quality (D) ยท 21 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 42.0 tok/s | Runs great |
| InternLM2 Math Plus 20B | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 29.1 tok/s | Runs well | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 21.3 tok/s | Runs well |
| Qwen2.5-Omni 3B | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 53.0 tok/s | Runs great | 29Basic quality (D) ยท 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 53.0 tok/s | Runs great |
| Unlimited OCR (3B-A0.6B) | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 114 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 95.9 tok/s | Runs great |
| StarCoder2 15B Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 36.2 tok/s | Runs great | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 26.5 tok/s | Runs well |
| Granite 3.2 2B Instruct | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 53.5 tok/s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 85.1 tok/s | Runs great |
| Granite 3.3 2B | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 53.5 tok/s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 85.1 tok/s | Runs great |
| EXAONE 3.5 2.4B | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 56.2 tok/s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 41.2 tok/s | Runs great |
| LLaVA 1.6 Mistral 7B | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 52.9 tok/s | Runs great | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 38.8 tok/s | Runs great |
| Zephyr 7B Beta | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Qwen3 1.7B | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 66.7 tok/s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 48.9 tok/s | Runs great |
| Yi 1.5 6B Chat | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Runs great | 27Basic quality (D) ยท 27 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 48.4 tok/s | Runs great |
| DeepSeek Coder 6.7B Instruct | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| LLM4Decompile 9B V2 | 37Basic quality (D) ยท 37 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 39.6 tok/s | Runs great |
| LFM2.5 1.2B Instruct | 32Basic quality (D) ยท 32 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 116 tok/s | Runs great | 32Basic quality (D) ยท 32 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 84.8 tok/s | Runs great |
| Granite 3.1 2B Instruct | 28Basic quality (D) ยท 28 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 53.5 tok/s | Runs great | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 85.1 tok/s | Runs great |
| Gemma 2 2B | 28Basic quality (D) ยท 28 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 51.9 tok/s | Runs great | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 82.5 tok/s | Runs great |
| CodeGemma 7B | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 46.9 tok/s | Runs great | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 40.9 tok/s | Runs great |
| MiniCPM-OCR 1.0 1.3B | 42Fair quality (C) ยท 42 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 104 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 76.3 tok/s | Runs great |
| EXAONE 4.0 1.2B | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 106 tok/s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 77.5 tok/s | Runs great |
| WeVisOne VL 2B | 27Basic quality (D) ยท 27 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 55.5 tok/s | Runs great | 27Basic quality (D) ยท 27 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 40.7 tok/s | Runs great |
| LLaVA 1.5 13B | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 43.2 tok/s | Runs great | 18Basic quality (D) ยท 18 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 40.9 tok/s | Runs great |
| Codestral Mamba 7B | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 40.3 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 40.2 tok/s | Runs great |
| LLM4Decompile 6.7B V2 | 32Basic quality (D) ยท 32 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| PaddleOCR-VL 1.6 0.9B | 42Fair quality (C) ยท 42 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 141 tok/s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 103 tok/s | Runs great |
| Granite Vision 3.2 2B | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 45.4 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 72.2 tok/s | Runs great |
| Code Llama 13B Instruct | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 44.3 tok/s | Runs great | 28Basic quality (D) ยท 28 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 32.5 tok/s | Runs great |
| WizardLM 13B v1.2 | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 44.3 tok/s | Runs great | 18Basic quality (D) ยท 18 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 42.0 tok/s | Runs great |
| TeleOCR | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 95.4 tok/s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at F16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 69.9 tok/s | Runs great |
| Mathstral 7B | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great | 33Basic quality (D) ยท 33 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Command A (111B) | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 7.3 tok/s | Slowover GPU memory limit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 5.3 tok/s | Slowover GPU memory limit |
| Granite 3.0 2B Instruct | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 51.5 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 81.8 tok/s | Runs great |
| Orca 2 7B | 22Basic quality (D) ยท 22 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 21Basic quality (D) ยท 21 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Qwen2.5 Coder 3B Instruct | 36Basic quality (D) ยท 36 of 100: how capable the model is at F16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 43.8 tok/s | Runs great | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 69.7 tok/s | Runs great |
| Qwen2.5-Math 7B Instruct | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 52.6 tok/s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 38.5 tok/s | Runs great |
| Granite Code 20B Instruct | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 28.8 tok/s | Runs well | 34Basic quality (D) ยท 34 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 21.1 tok/s | Runs well |
| Granite 4.0 H 1B | 28Basic quality (D) ยท 28 of 100: how capable the model is at F16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 92.7 tok/s | Runs great | 28Basic quality (D) ยท 28 of 100: how capable the model is at F16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 67.9 tok/s | Runs great |
| SQLCoder 7B v2 | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 29Basic quality (D) ยท 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Vicuna 13B v1.5 | 19Basic quality (D) ยท 19 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 44.3 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 42.0 tok/s | Runs great |
| Laguna S 2.1 (118B-A8B) | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 73.5 tok/s | Slowfar over GPU memory limit | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 58.0 tok/s | Slowfar over GPU memory limit |
| SmolVLM2 2.2B Instruct | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 60.2 tok/s | Runs great | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 44.1 tok/s | Runs great |
| Llama 2 7B Chat | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 19Basic quality (D) ยท 19 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Falcon-H1 1.5B Instruct | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 87.4 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 64.0 tok/s | Runs great |
| Ling 3.0 flash Sante | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 99.9 tok/s | Slowfar over GPU memory limit | 47Fair quality (C) ยท 47 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 81.2 tok/s | Slowfar over GPU memory limit |
| Qwen2-VL 2B Instruct | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 61.3 tok/s | Runs great | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 44.9 tok/s | Runs great |
| Granite Code 8B Instruct (128K) | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 49.7 tok/s | Runs great | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 43.4 tok/s | Runs great |
| StarCoder2 15B (base) | 30Basic quality (D) ยท 30 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 36.2 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 34.2 tok/s | Runs great |
| Baichuan 2 7B Chat | 19Basic quality (D) ยท 19 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 53.3 tok/s | Runs great | 19Basic quality (D) ยท 19 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 39.1 tok/s | Runs great |
| LFM2 1.2B | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 116 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 84.8 tok/s | Runs great |
| SmolVLM 2.2B Instruct | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 60.2 tok/s | Runs great | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 44.1 tok/s | Runs great |
| InternLM2 Math Plus 7B | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 51.7 tok/s | Runs great | 31Basic quality (D) ยท 31 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 45.1 tok/s | Runs great |
| Zamba2 2.7B Instruct | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 50.9 tok/s | Runs great | 22Basic quality (D) ยท 22 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 80.9 tok/s | Runs great |
| DeepSeek LLM 7B Chat | 18Basic quality (D) ยท 18 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 42.5 tok/s | Runs great | 17Basic quality (D) ยท 17 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 42.5 tok/s | Runs great |
| Qwen3.5 0.8B | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 156 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 114 tok/s | Runs great |
| Granite 3.1 3B-A800M Instruct | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 97.2 tok/s | Runs great | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 80.4 tok/s | Runs great |
| Qwen2.5 1.5B Instruct | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 87.9 tok/s | Runs great | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 64.4 tok/s | Runs great |
| LFM2-VL 1.6B | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 85.7 tok/s | Runs great | 24Basic quality (D) ยท 24 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 62.8 tok/s | Runs great |
| Nemotron Mini 4B Instruct | 22Basic quality (D) ยท 22 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 70.1 tok/s | Runs great | 22Basic quality (D) ยท 22 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 51.4 tok/s | Runs great |
| Yi 6B Chat | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 48.5 tok/s | Runs great | 19Basic quality (D) ยท 19 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 48.4 tok/s | Runs great |
| Hunyuan 1.8B Instruct | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 75.6 tok/s | Runs great | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 55.4 tok/s | Runs great |
| LLaVA 1.5 7B | 16Basic quality (D) ยท 16 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 41.6 tok/s | Runs great | 15Basic quality (D) ยท 15 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 41.6 tok/s | Runs great |
| Command R+ (08-2024) 104B | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 7.8 tok/s | Slowover GPU memory limit | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 5.7 tok/s | Slowover GPU memory limit |
| Granite 3.0 3B-A800M Instruct | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 97.2 tok/s | Runs great | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 80.4 tok/s | Runs great |
| SmolLM2 1.7B Instruct | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 79.2 tok/s | Runs great | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 58.0 tok/s | Runs great |
| Code Llama 7B Instruct | 24Basic quality (D) ยท 24 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 23Basic quality (D) ยท 23 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| Llama 3.2 1B Instruct | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 109 tok/s | Runs great | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 80.0 tok/s | Runs great |
| WizardMath 7B v1.1 | 26Basic quality (D) ยท 26 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great | 25Basic quality (D) ยท 25 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 40.5 tok/s | Runs great |
| Vicuna 7B v1.5 | 14Basic quality (D) ยท 14 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 43.6 tok/s | Runs great | 13Basic quality (D) ยท 13 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 43.5 tok/s | Runs great |
| StableLM Zephyr 3B | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 48.4 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 76.9 tok/s | Runs great |
| OLMo 7B Instruct | 14Basic quality (D) ยท 14 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great | 13Basic quality (D) ยท 13 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 42.6 tok/s | Runs great |
| InternLM2.5 1.8B Chat | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 71.6 tok/s | Runs great | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 52.5 tok/s | Runs great |
| Phi-2 2.7B | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 48.7 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 77.4 tok/s | Runs great |
| StarCoder 15B | 24Basic quality (D) ยท 24 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 36.5 tok/s | Runs great | 21Basic quality (D) ยท 21 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 34.6 tok/s | Runs great |
| Command A Reasoning (111B) | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 7.3 tok/s | Slowover GPU memory limit | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 5.3 tok/s | Slowover GPU memory limit |
| Gemma 3 1B | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 135 tok/s | Runs great | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 99.2 tok/s | Runs great |
| StarCoder2 7B | 24Basic quality (D) ยท 24 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 41.0 tok/s | Runs great | 23Basic quality (D) ยท 23 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 40.9 tok/s | Runs great |
| InternLM 7B | 12Basic quality (D) ยท 12 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 40.1 tok/s | Runs great | 11Basic quality (D) ยท 11 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 40.1 tok/s | Runs great |
| Command R+ 104B (04-2024) | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 7.8 tok/s | Slowover GPU memory limit | 44Fair quality (C) ยท 44 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 5.7 tok/s | Slowover GPU memory limit |
| GLiDE 2B | 17Basic quality (D) ยท 17 of 100: how capable the model is at F16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 72.0 tok/s | Runs great | 17Basic quality (D) ยท 17 of 100: how capable the model is at F16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 52.8 tok/s | Runs great |
| Falcon 40B Instruct | 11Basic quality (D) ยท 11 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 22.7 tok/s | Runs well | 11Basic quality (D) ยท 11 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 16.6 tok/s | Runs well |
| Falcon 3 1B Instruct | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 81.1 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 59.4 tok/s | Runs great |
| Spark X2.6 1.7B | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 79.2 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 58.0 tok/s | Runs great |
| LFM2 700M | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 183 tok/s | Runs great | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 134 tok/s | Runs great |
| Qwen2 1.5B Instruct | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 87.9 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 64.4 tok/s | Runs great |
| DeepScaleR 1.5B Preview | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 76.1 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 55.7 tok/s | Runs great |
| Qwen2.5 Coder 1.5B Instruct | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 87.9 tok/s | Runs great | 26Basic quality (D) ยท 26 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 64.4 tok/s | Runs great |
| Mistral Medium 3.5 (128B) | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 6.3 tok/s | Slowfar over GPU memory limit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 4.6 tok/s | Slowfar over GPU memory limit |
| Mistral Large 2.1 (2411) | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 6.6 tok/s | Slowfar over GPU memory limit | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 4.8 tok/s | Slowfar over GPU memory limit |
| SmolLM 1.7B Instruct | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 79.2 tok/s | Runs great | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 58.0 tok/s | Runs great |
| Stable Code Instruct 3B | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 48.4 tok/s | Runs great | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 76.9 tok/s | Runs great |
| OLMo 2 1B Instruct | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 91.5 tok/s | Runs great | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 67.0 tok/s | Runs great |
| Gemma 2B IT | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. | 53.9 tok/s | Runs great | 12Basic quality (D) ยท 12 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 85.8 tok/s | Runs great |
| Mistral Large 2 (123B) | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 6.6 tok/s | Slowfar over GPU memory limit | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 4.8 tok/s | Slowfar over GPU memory limit |
| Pixtral Large 124B (2411) | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 6.5 tok/s | Slowfar over GPU memory limit | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 4.8 tok/s | Slowfar over GPU memory limit |
| Qwen3 0.6B | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 181 tok/s | Runs great | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 132 tok/s | Runs great |
| Granite Code 3B Instruct (128K) | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 84.4 tok/s | Runs great | 20Basic quality (D) ยท 20 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 61.9 tok/s | Runs great |
| Falcon-H1 0.5B Instruct | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 260 tok/s | Runs great | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 191 tok/s | Runs great |
| DeepSeek R1 Distill Qwen 1.5B | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 76.1 tok/s | Runs great | 22Basic quality (D) ยท 22 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 55.7 tok/s | Runs great |
| LFM2.5 350M | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 387 tok/s | Runs great | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 283 tok/s | Runs great |
| StableLM 2 1.6B Chat | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 82.6 tok/s | Runs great | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 60.5 tok/s | Runs great |
| StableLM 2 Zephyr 1.6B | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 82.6 tok/s | Runs great | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 60.5 tok/s | Runs great |
| Falcon 7B Instruct | 6Basic quality (D) ยท 6 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 40.7 tok/s | Runs great | 5Basic quality (D) ยท 5 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 40.6 tok/s | Runs great |
| d1 2B | 21Basic quality (D) ยท 21 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 70.5 tok/s | Runs great | 21Basic quality (D) ยท 21 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 51.7 tok/s | Runs great |
| StarCoder2 3B | 18Basic quality (D) ยท 18 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 44.7 tok/s | Runs great | 18Basic quality (D) ยท 18 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 71.0 tok/s | Runs great |
| LFM2-VL 450M | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 301 tok/s | Runs great | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | 220 tok/s | Runs great |
| LFM2 350M | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 387 tok/s | Runs great | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 283 tok/s | Runs great |
| Granite 4.0 H 350M | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 398 tok/s | Runs great | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 292 tok/s | Runs great |
| GLiDE 0.8B | 11Basic quality (D) ยท 11 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 181 tok/s | Runs great | 11Basic quality (D) ยท 11 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 132 tok/s | Runs great |
| Qwen1.5 110B Chat | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 7.2 tok/s | Slowover GPU memory limit | 36Basic quality (D) ยท 36 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 5.3 tok/s | Slowover GPU memory limit |
| Qwen2.5-Math 1.5B Instruct | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 87.9 tok/s | Runs great | 20Basic quality (D) ยท 20 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 64.4 tok/s | Runs great |
| SmolVLM 500M Instruct | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 266 tok/s | Runs great | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 195 tok/s | Runs great |
| Qwen2.5 0.5B Instruct | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 276 tok/s | Runs great | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 202 tok/s | Runs great |
| Devstral 2 (123B) | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking. | 6.4 tok/s | Slowfar over GPU memory limit | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 4.7 tok/s | Slowfar over GPU memory limit |
| LFM2.5 230M | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 589 tok/s | Runs great | 10Basic quality (D) ยท 10 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 431 tok/s | Runs great |
| Hunyuan 0.5B Instruct | 8Basic quality (D) ยท 8 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 251 tok/s | Runs great | 8Basic quality (D) ยท 8 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 184 tok/s | Runs great |
| Yi Coder 1.5B Chat | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 91.5 tok/s | Runs great | 16Basic quality (D) ยท 16 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 67.0 tok/s | Runs great |
| SmolLM2 360M Instruct | 8Basic quality (D) ยท 8 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 376 tok/s | Runs great | 8Basic quality (D) ยท 8 of 100: how capable the model is at F16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking. | 276 tok/s | Runs great |
| CodeGemma 2B | 14Basic quality (D) ยท 14 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 53.9 tok/s | Runs great | 14Basic quality (D) ยท 14 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 85.8 tok/s | Runs great |
| TinyLlama 1.1B Chat | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 123 tok/s | Runs great | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 90.2 tok/s | Runs great |
| Qwen2 0.5B Instruct | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 276 tok/s | Runs great | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 202 tok/s | Runs great |
| ERNIE 4.5 0.3B | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 376 tok/s | Runs great | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 276 tok/s | Runs great |
| SmolVLM 256M Instruct | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 521 tok/s | Runs great | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 382 tok/s | Runs great |
| Gemma 3 270M | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 502 tok/s | Runs great | 6Basic quality (D) ยท 6 of 100: how capable the model is at F16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | 367 tok/s | Runs great |
| SmolLM 360M Instruct | 5Basic quality (D) ยท 5 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 376 tok/s | Runs great | 5Basic quality (D) ยท 5 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 276 tok/s | Runs great |
| DeepSeek Coder 1.3B Instruct | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 100 tok/s | Runs great | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 73.5 tok/s | Runs great |
| LLM4Decompile 1.3B V2 | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 100 tok/s | Runs great | 12Basic quality (D) ยท 12 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 73.5 tok/s | Runs great |
| surya ocr 2.5 | 15Basic quality (D) ยท 15 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 196 tok/s | Runs great | 15Basic quality (D) ยท 15 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 144 tok/s | Runs great |
| SmolLM2 135M Instruct | 4Basic quality (D) ยท 4 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 1,042 tok/s | Runs great | 4Basic quality (D) ยท 4 of 100: how capable the model is at F16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | 763 tok/s | Runs great |
| Falcon-H1 Tiny 90M Instruct | 3Basic quality (D) ยท 3 of 100: how capable the model is at F16, whatever it runs on. Match score 48/100 (quality, fit, speed and memory use) sets the ranking. | 1,505 tok/s | Runs great | 3Basic quality (D) ยท 3 of 100: how capable the model is at F16, whatever it runs on. Match score 48/100 (quality, fit, speed and memory use) sets the ranking. | 1,102 tok/s | Runs great |
| SmolLM 135M Instruct | 2Basic quality (D) ยท 2 of 100: how capable the model is at F16, whatever it runs on. Match score 48/100 (quality, fit, speed and memory use) sets the ranking. | 1,042 tok/s | Runs great | 2Basic quality (D) ยท 2 of 100: how capable the model is at F16, whatever it runs on. Match score 48/100 (quality, fit, speed and memory use) sets the ranking. | 763 tok/s | Runs great |
| Qwen2.5 Coder 0.5B Instruct | 8Basic quality (D) ยท 8 of 100: how capable the model is at F16, whatever it runs on. Match score 46/100 (quality, fit, speed and memory use) sets the ranking. | 276 tok/s | Runs great | 8Basic quality (D) ยท 8 of 100: how capable the model is at F16, whatever it runs on. Match score 46/100 (quality, fit, speed and memory use) sets the ranking. | 202 tok/s | Runs great |
| gpt-oss 120B | 84Great quality (A) ยท 84 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 84Great quality (A) ยท 84 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K3.1 Pro (2.8T) | 69Good quality (B) ยท 69 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 69Good quality (B) ยท 69 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V4.1 Flash 1017 (552B-A16B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-5.3 (753B-A40B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3.8 Max (2.4T-A95B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K3 (2.8T) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| LongCat 2.5 (1.6T-A48B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiMo V2.6 Pro (1T-A42B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Ling 3.1 Flash 1019 (560B-A25B) | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 68Good quality (B) ยท 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax M3.1 Flash (428B-A23B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Nex N2.6 (321B-A18B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Step 5 Preview (600B-A27B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-5.3 Flash (320B-A18B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiMo V2.6 Flash (310B-A15B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V4.1 Flash (552B-A16B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Mistral Large 4 (1T-A52B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3.8 Flash Next (125B-A6B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-5.2 (753B-A40B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Hunyuan Hy4 Preview (780B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Muse Spark Open (816B-A42B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V4 Pro (1.6T-A49B) | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 67Good quality (B) ยท 67 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K2.6 (1T) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Ling 3.1 Flash (560B-A25B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiMo V2.5 Pro (1T-A42B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Intern-S2 (397B-A17B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 9/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-5.1 (744B-A40B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Nex N2 Pro (397B-A17B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V4 Pro (April 2026) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Naive N0.5 Flash (309B-A15.5B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3.5 397B-A17B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax M3 (427B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-5 (744B-A40B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| K2-Horizon (375B-A23B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Hunyuan Hy3 (295B-A21B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Beam 1.0 (501B-A23B) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K2.5 (1T) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K2.7 Code (1T) | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 66Good quality (B) ยท 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V4 Flash (284B-A13B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiMo V2.5 (310B-A15B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Ornith 1.5 397B-A17B | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V3.2 (671B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Inkling Small (276B-A12B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-4.7 (355B-A32B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Inkling (975B-A41B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax M2.7 (230B-A10B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Step 3.7 Flash (198B-A11B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Nemotron 3 Ultra 550B-A55B | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek R1 0528 (671B) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K2 Thinking (1T) | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 65Good quality (B) ยท 65 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V3.2 Exp | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Ling 2.6 1T | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax M2.5 (230B-A10B) | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek R1 (671B) | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V4 Flash (April 2026) | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Step 3.5 Flash (196B-A11B) | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V3.1 Terminus | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 64Good quality (B) ยท 64 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V3.1 (671B) | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K2 Instruct 0905 (1T) | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax M2.1 (230B-A10B) | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3 235B-A22B Thinking 2507 | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3 Coder 480B-A35B | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiMo V2 Flash (309B-A15B) | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Ling 1T | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Mistral Large 3 (675B) | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3 235B-A22B Instruct 2507 | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Solar Open 2 250B | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 63Good quality (B) ยท 63 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Kimi K2 Instruct (July 2025) | 62Good quality (B) ยท 62 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 62Good quality (B) ยท 62 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax M2 (230B-A10B) | 62Good quality (B) ยท 62 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 62Good quality (B) ยท 62 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3-VL 235B-A22B Instruct | 62Good quality (B) ยท 62 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 62Good quality (B) ยท 62 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Command A+ (218B-A25B) | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| K-EXAONE 236B-A23B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| LongCat Flash Chat (560B) | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Qwen3 235B-A22B | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 61Good quality (B) ยท 61 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| InternVL3.5 241B-A28B | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Llama 3.1 Nemotron Ultra 253B v1 | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 60Good quality (B) ยท 60 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 8/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Baichuan M3 235B-A22B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-4.6 (355B-A32B) | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V3 0324 (671B) | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax-M1 80K | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 58Good quality (B) ยท 58 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| GLM-4.5 (355B-A32B) | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax-M1 40K | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 57Good quality (B) ยท 57 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Trinity Large Thinking (400B-A13B) | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| ERNIE 4.5 300B-A47B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Hermes 4 405B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 56Good quality (B) ยท 56 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| NorthCUA Pro | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 54Fair quality (C) ยท 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| dots.llm1 Instruct | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| MiniMax-Text-01 | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Tรผlu 3 405B | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 53Fair quality (C) ยท 53 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V3 (671B) | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 52Fair quality (C) ยท 52 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 7/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek Coder V2 (236B) | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 51Fair quality (C) ยท 51 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek Coder V2 Instruct (0614) | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V2.5 1210 | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Llama 3.1 405B Instruct | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Llama 4 Maverick (17B x 128E) | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 50Fair quality (C) ยท 50 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V2.5 | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Jamba 1.5 Large | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 49Fair quality (C) ยท 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 6/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| DeepSeek V2 Chat | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 43Fair quality (C) ยท 43 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Mixtral 8x22B Instruct | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 38Basic quality (D) ยท 38 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Grok-1 (314B) | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 4/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 4/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Snowflake Arctic Instruct | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 4/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 35Basic quality (D) ยท 35 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 4/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
| Falcon 180B Chat | 23Basic quality (D) ยท 23 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 3/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 23Basic quality (D) ยท 23 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 3/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
Image models
61 of 61 image models run on the Apple M4 Max (64 GB). 48.0 GB usable for models.
61 of 61 image models run on the Apple M1 Max (64 GB). 48.0 GB usable for models.
| Model | QualityM4 Max | SpeedM4 Max | RunsM4 Max | QualityM1 Max | SpeedM1 Max | RunsM1 Max |
|---|---|---|---|---|---|---|
| FLUX.2 [klein] 9B | 84Great quality (A) ยท 84 of 100: how capable the model is at FP16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 11 s | Runs great | 84Great quality (A) ยท 84 of 100: how capable the model is at FP16, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 16 s | Runs well |
| Krea 2 Turbo | 92Excellent quality (A+) ยท 92 of 100: how capable the model is at FP16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 25 s | Runs well | 92Excellent quality (A+) ยท 92 of 100: how capable the model is at FP16, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 39 s | Runs well |
| ERNIE-Image Turbo | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at FP16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 19 s | Runs well | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at FP16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 29 s | Runs well |
| Z-Image Turbo | 84Great quality (A) ยท 84 of 100: how capable the model is at FP16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 15 s | Runs well | 84Great quality (A) ยท 84 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 22 s | Runs well |
| Qwen-Image 2.1 Turbo | 87Excellent quality (A+) ยท 87 of 100: how capable the model is at FP8, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 61 s | Slow | 87Excellent quality (A+) ยท 87 of 100: how capable the model is at FP8, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 min | Slow |
| FLUX 3 Image Klein | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at FP16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 33 s | Runs well | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 51 s | Slow |
| Fibo 2.0 Dev | 93Excellent quality (A+) ยท 93 of 100: how capable the model is at FP16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 2.2 min | Slow | 93Excellent quality (A+) ยท 93 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 3.4 min | Very slow |
| Ideogram 4 | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at FP8, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 2.2 min | Slow | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at FP8, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 3.4 min | Very slow |
| FLUX.2 [klein] 4B | 78Great quality (A) ยท 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 5.9 s | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 9.0 s | Runs great |
| ERNIE-Image | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 2.1 min | Slow | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 3.3 min | Very slow |
| Ming-Image 0.1 Design | 84Great quality (A) ยท 84 of 100: how capable the model is at FP8, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 86 s | Slow | 84Great quality (A) ยท 84 of 100: how capable the model is at FP8, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 2.2 min | Slow |
| FLUX.1 Krea [dev] | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 70 s | Slow | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 1.8 min | Slow |
| MiniMax H3 Image | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at FP8, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 55 s | Slowencoder unloaded after use | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at FP8, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 85 s | Slowencoder unloaded after use |
| Stable Diffusion 3.5 Large Turbo | 62Good quality (B) ยท 62 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 9.4 s | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 13 s | Runs great |
| Qwen-Image 2512 | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 10.1 min | Very slow | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 15.5 min | Very slow |
| Qwen-Image 2.1 | 92Excellent quality (A+) ยท 92 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 6.7 min | Very slowtight fit | 92Excellent quality (A+) ยท 92 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 10.3 min | Very slowtight fit |
| FLUX.1 [dev] | 80Great quality (A) ยท 80 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 70 s | Slow | 80Great quality (A) ยท 80 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 1.8 min | Slow |
| MOSS-Image SFT | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 82 s | Slow | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 2.1 min | Slow |
| FLUX.1 Kontext [dev] | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 2.1 min | Slow | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 3.3 min | Very slow |
| Qwen-Image | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 7.9 min | Very slow | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 12.1 min | Very slow |
| FireRed-Image-Edit 1.1 | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 9.0 min | Very slow | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 13.8 min | Very slow |
| Qwen-Image-Edit 2511 | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 9.0 min | Very slow | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 13.8 min | Very slow |
| HiDream-I1 Dev | 78Great quality (A) ยท 78 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 90 s | Slow | 78Great quality (A) ยท 78 of 100: how capable the model is at FP8, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 2.3 min | Very slow |
| HiDream-I1 Full | 83Great quality (A) ยท 83 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 5.1 min | Very slow | 83Great quality (A) ยท 83 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 7.8 min | Very slow |
| HiDream-I1 Fast | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 53 s | Slow | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 81 s | Slow |
| FLUX.1 [schnell] | 68Good quality (B) ยท 68 of 100: how capable the model is at FP8, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 11 s | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 18 s | Runs well |
| FLUX.1 Fill [dev] | 76Great quality (A) ยท 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 73 s | Slow | 76Great quality (A) ยท 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 1.9 min | Slow |
| FLUX.2 [dev] | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 4.2 min | Very slowencoder unloaded after use | 91Excellent quality (A+) ยท 91 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 6.5 min | Very slowencoder unloaded after use |
| Chroma1-HD | 78Great quality (A) ยท 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 2.0 min | Slow | 78Great quality (A) ยท 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 3.0 min | Very slow |
| Z-Image | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 1.7 min | Slow | 82Great quality (A) ยท 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 2.5 min | Very slow |
| CogView4 6B | 70Great quality (A) ยท 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 1.8 min | Slow | 70Great quality (A) ยท 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 2.7 min | Very slow |
| Stable Diffusion 3.5 Large | 66Good quality (B) ยท 66 of 100: how capable the model is at FP16, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 64 s | Slow | 66Good quality (B) ยท 66 of 100: how capable the model is at FP16, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 min | Slow |
| Kolors | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 23 s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 36 s | Runs well |
| Iris-3B | 72Great quality (A) ยท 72 of 100: how capable the model is at FP16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 min | Slow | 72Great quality (A) ยท 72 of 100: how capable the model is at FP16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 2.4 min | Very slow |
| DreamShaper XL v2 Turbo | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 7.6 s | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 11 s | Runs great |
| HunyuanImage 3.0 (80B-A13B) | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP8, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 6.9 min | Very slowCPU offload | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP8, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 10.5 min | Very slowCPU offload |
| PixArt-ฮฃ XL 1024 | 50Fair quality (C) ยท 50 of 100: how capable the model is at FP16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 8.8 s | Runs great | 50Fair quality (C) ยท 50 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 13 s | Runs well |
| SDXL-Lightning (4-step) | 54Fair quality (C) ยท 54 of 100: how capable the model is at FP16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 4.1 s | Runs great | 54Fair quality (C) ยท 54 of 100: how capable the model is at FP16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 6.3 s | Runs great |
| OmniGen2 | 66Good quality (B) ยท 66 of 100: how capable the model is at FP16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 min | Slow | 64Good quality (B) ยท 64 of 100: how capable the model is at FP8, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 2.2 min | Slow |
| Cosmos3-Super Text2Image (64B) | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 4.6 min | Very slowCPU offload | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 7.0 min | Very slowCPU offload |
| Stable Diffusion 3.5 Medium | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 35 s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 54 s | Slow |
| Juggernaut XL v9 | 62Good quality (B) ยท 62 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 26 s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at FP8, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 36 s | Runs well |
| RealVisXL V5.0 | 62Good quality (B) ยท 62 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 26 s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at FP8, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 36 s | Runs well |
| Animagine XL 4.0 | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 25 s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 38 s | Runs well |
| Illustrious XL v0.1 | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 25 s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 38 s | Runs well |
| Kandinsky 3 | 48Fair quality (C) ยท 48 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 53 s | Slow | 48Fair quality (C) ยท 48 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 81 s | Slow |
| AuraFlow v0.3 | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 82 s | Slow | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 2.1 min | Slow |
| Stable Diffusion 3 Medium | 50Fair quality (C) ยท 50 of 100: how capable the model is at FP8, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 16 s | Runs well | 50Fair quality (C) ยท 50 of 100: how capable the model is at FP8, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 25 s | Runs well |
| Pony Diffusion V6 XL | 56Good quality (B) ยท 56 of 100: how capable the model is at FP8, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 20 s | Runs well | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 34 s | Runs well |
| Lumina-Image 2.0 | 60Good quality (B) ยท 60 of 100: how capable the model is at FP8, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 37 s | Runs well | 62Good quality (B) ยท 62 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 63 s | Slow |
| Animagine XL 3.1 | 54Fair quality (C) ยท 54 of 100: how capable the model is at FP8, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 22 s | Runs well | 56Good quality (B) ยท 56 of 100: how capable the model is at FP16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 38 s | Runs well |
| Playground v2.5 | 58Good quality (B) ยท 58 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 40 s | Runs well | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 67 s | Slow |
| SDXL Turbo | 40Fair quality (C) ยท 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 0.9 s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 1.3 s | Runs great |
| Stable Diffusion XL 1.0 | 53Fair quality (C) ยท 53 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 24 s | Runs well | 53Fair quality (C) ยท 53 of 100: how capable the model is at FP8, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 36 s | Runs well |
| Stable Cascade | 50Fair quality (C) ยท 50 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 26 s | Runs well | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | 45 s | Slow |
| Hunyuan-DiT v1.2 | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 35 s | Runs well | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 54 s | Slow |
| Realistic Vision V5.1 | 42Fair quality (C) ยท 42 of 100: how capable the model is at FP16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | 7.0 s | Runs great | 42Fair quality (C) ยท 42 of 100: how capable the model is at FP16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 11 s | Runs great |
| DreamShaper 8 | 40Fair quality (C) ยท 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 7.0 s | Runs great | 40Fair quality (C) ยท 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 11 s | Runs great |
| Stable Diffusion 1.5 | 35Basic quality (D) ยท 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking. | 7.0 s | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking. | 11 s | Runs great |
| LCM DreamShaper v7 | 30Basic quality (D) ยท 30 of 100: how capable the model is at FP16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 1.5 s | Runs great | 30Basic quality (D) ยท 30 of 100: how capable the model is at FP16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 2.2 s | Runs great |
| SD Turbo | 28Basic quality (D) ยท 28 of 100: how capable the model is at FP16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 0.5 s | Runs great | 28Basic quality (D) ยท 28 of 100: how capable the model is at FP16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking. | 0.7 s | Runs great |
Video models
42 of 43 video models run on the Apple M4 Max (64 GB). 48.0 GB usable for models.
42 of 43 video models run on the Apple M1 Max (64 GB). 48.0 GB usable for models.
| Model | QualityM4 Max | SpeedM4 Max | RunsM4 Max | QualityM1 Max | SpeedM1 Max | RunsM1 Max |
|---|---|---|---|---|---|---|
| LTX-2.5 (22B, video + audio) | 81Great quality (A) ยท 81 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 10.7 min | Runs well | 81Great quality (A) ยท 81 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 16.4 min | Slow |
| LTX-Video 13B 0.9.8 (distilled) | 66Good quality (B) ยท 66 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 5.3 min | Runs well | 66Good quality (B) ยท 66 of 100: how capable the model is at FP8, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 8.1 min | Runs well |
| LTX-2.3 | 84Great quality (A) ยท 84 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 35.2 min | Slowtight fit | 84Great quality (A) ยท 84 of 100: how capable the model is at FP8, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 53.9 min | Very slowtight fit |
| Wan 2.2 I2V A14B | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 3.0 h | Very slow | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 4.6 h | Very slow |
| Wan 2.2 T2V A14B | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 3.0 h | Very slow | 85Excellent quality (A+) ยท 85 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 4.6 h | Very slow |
| HunyuanVideo 1.5 (8.3B) | 82Great quality (A) ยท 82 of 100: how capable the model is at FP8, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 2.6 h | Very slow | 82Great quality (A) ยท 82 of 100: how capable the model is at FP8, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 4.0 h | Very slow |
| LTX-2 (19B, video + audio) | 80Great quality (A) ยท 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 35.2 min | Slowtight fit | 80Great quality (A) ยท 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 53.9 min | Very slowtight fit |
| Bernini-R A14B (video generation + editing) | 83Great quality (A) ยท 83 of 100: how capable the model is at FP8, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 63.3 min | Very slow | 83Great quality (A) ยท 83 of 100: how capable the model is at FP8, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 h | Very slow |
| SCAIL-2 14B (character animation) | 78Great quality (A) ยท 78 of 100: how capable the model is at FP8, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 1.8 h | Very slow | 78Great quality (A) ยท 78 of 100: how capable the model is at FP8, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 2.7 h | Very slow |
| Wan 2.1 I2V 14B 720P | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 3.7 h | Very slow | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 5.7 h | Very slow |
| Wan 2.1 T2V 14B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 3.7 h | Very slow | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 5.7 h | Very slow |
| FramePack I2V (HunyuanVideo 13B) | 72Great quality (A) ยท 72 of 100: how capable the model is at FP8, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 35.2 min | Slow | 72Great quality (A) ยท 72 of 100: how capable the model is at FP8, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 53.9 min | Very slow |
| MiniMax H3 (33B, video + audio) | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 2.0 h | Very slowtight fit ยท encoder unloaded after use | 89Excellent quality (A+) ยท 89 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 3.1 h | Very slowtight fit ยท encoder unloaded after use |
| Wan 2.2 S2V 14B (speech to video) | 75Great quality (A) ยท 75 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 1.9 h | Very slow | 75Great quality (A) ยท 75 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 2.9 h | Very slow |
| Kandinsky 6.0 Video Lite (3B, video + audio) | 68Good quality (B) ยท 68 of 100: how capable the model is at FP16, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 30.3 min | Slow | 68Good quality (B) ยท 68 of 100: how capable the model is at FP16, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 46.4 min | Very slow |
| Wan 2.2 TI2V 5B | 68Good quality (B) ยท 68 of 100: how capable the model is at FP8, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 45.7 min | Very slow | 68Good quality (B) ยท 68 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 70.1 min | Very slow |
| Wan 2.2 Animate 14B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 3.6 h | Very slowtight fit | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 5.5 h | Very slowtight fit |
| Wan 2.1 VACE 14B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 4.1 h | Very slowtight fit | 77Great quality (A) ยท 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 6.3 h | Very slowtight fit |
| SkyReels V2 T2V 14B 720P | 76Great quality (A) ยท 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 3.5 h | Very slowtight fit | 76Great quality (A) ยท 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 5.4 h | Very slowtight fit |
| HY-OmniWeaving (8.3B, unified video) | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 46.6 min | Very slow | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 71.5 min | Very slow |
| SkyReels V2 T2V 14B 540P | 72Great quality (A) ยท 72 of 100: how capable the model is at FP8, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 2.1 h | Very slow | 72Great quality (A) ยท 72 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 3.1 h | Very slow |
| Wan 2.1 I2V 14B 480P | 72Great quality (A) ยท 72 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 79.1 min | Very slow | 73Great quality (A) ยท 73 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 2.6 h | Very slow |
| Cosmos-Predict2 14B Video2World | 68Good quality (B) ยท 68 of 100: how capable the model is at FP8, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 2.3 h | Very slow | 68Good quality (B) ยท 68 of 100: how capable the model is at FP8, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 3.6 h | Very slow |
| HunyuanVideo 13B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 4.7 h | Very slow | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 7.2 h | Very slow |
| HunyuanVideo I2V 13B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 4.7 h | Very slow | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 7.2 h | Very slow |
| Cosmos-Predict2 2B Video2World | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 24.4 min | Slow | 58Good quality (B) ยท 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 37.5 min | Slow |
| Open-Sora 2.0 (11B) | 70Great quality (A) ยท 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 4.9 h | Very slowtight fit | 70Great quality (A) ยท 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 7.5 h | Very slowtight fit |
| LTX-Video 2B | 50Fair quality (C) ยท 50 of 100: how capable the model is at FP16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 2.4 min | Runs great | 49Fair quality (C) ยท 49 of 100: how capable the model is at FP8, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 3.4 min | Runs great |
| CogVideoX 1.5 5B | 58Good quality (B) ยท 58 of 100: how capable the model is at FP8, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | 79.1 min | Very slow | 58Good quality (B) ยท 58 of 100: how capable the model is at FP8, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 2.0 h | Very slow |
| CogVideoX 5B I2V | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 min | Slow | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 27.0 min | Slow |
| Step-Video T2V (30B) | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 4.8 h | Very slowtight fit ยท encoder unloaded after use | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 7.3 h | Very slowtight fit ยท encoder unloaded after use |
| Mochi 1 Preview (10B) | 60Good quality (B) ยท 60 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 h | Very slow | 60Good quality (B) ยท 60 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 2.5 h | Very slow |
| CogVideoX 5B | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 17.6 min | Slow | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 27.0 min | Slow |
| Kandinsky 6.0 Pro (29B, video + audio) | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 6.2 h | Very slowCPU offload | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 9.5 h | Very slowCPU offload |
| Allegro (2.8B) | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 2.0 h | Very slow | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 3.0 h | Very slow |
| Cosmos 1.0 Diffusion 7B Text2World | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 56.4 min | Very slowencoder unloaded after use | 60Good quality (B) ยท 60 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 86.5 min | Very slowencoder unloaded after use |
| Wan 2.1 VACE 1.3B | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 25.4 min | Slow | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 39.0 min | Slow |
| Wan 2.1 T2V 1.3B | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 23.4 min | Slow | 55Good quality (B) ยท 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 36.0 min | Slow |
| SkyReels V2 Diffusion Forcing 1.3B | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | 29.3 min | Slow | 52Fair quality (C) ยท 52 of 100: how capable the model is at FP16, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. | 44.9 min | Very slow |
| Wan-Dancer 14B (music to dance) | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 8.4 h | Very slowCPU offload | 74Great quality (A) ยท 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 12.9 h | Very slowCPU offload |
| CogVideoX 2B | 44Fair quality (C) ยท 44 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | 5.3 min | Runs well | 44Fair quality (C) ยท 44 of 100: how capable the model is at FP8, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | 8.1 min | Runs well |
| AnimateDiff v3 (SD 1.5) | 35Basic quality (D) ยท 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 2.0 min | Runs great | 35Basic quality (D) ยท 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking. | 3.0 min | Runs great |
| Cosmos3-Super Image2Video (64B) | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
Decision models
69 of 70 decision models run on the Apple M4 Max (64 GB). 48.0 GB usable for models.
69 of 70 decision models run on the Apple M1 Max (64 GB). 48.0 GB usable for models.
| Model | QualityM4 Max | SpeedM4 Max | RunsM4 Max | QualityM1 Max | SpeedM1 Max | RunsM1 Max |
|---|---|---|---|---|---|---|
| Jina Reranker v3.5 | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at F16, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 35 ms | Runs great | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at F16, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking. | 58 ms | Runs great |
| Jina Reranker v3 | 84Great quality (A) ยท 84 of 100: how capable the model is at F16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 35 ms | Runs great | 84Great quality (A) ยท 84 of 100: how capable the model is at F16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. | 58 ms | Runs great |
| d1 3B | 82Great quality (A) ยท 82 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 97 ms | Runs great | 82Great quality (A) ยท 82 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 142 ms | Runs well |
| mxbai-rerank-large v2 | 82Great quality (A) ยท 82 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 84 ms | Runs great | 82Great quality (A) ยท 82 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 133 ms | Runs well |
| Intern-Decision 4B | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 121 ms | Runs great | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 205 ms | Runs well |
| Qwen3 Reranker 4B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 198 ms | Runs well | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 342 ms | Runs well |
| FRIDA-Decisions | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 27 ms | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 45 ms | Runs great |
| GLiNER2.5 Decide | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 18 ms | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 28 ms | Runs great |
| Qwen3 Reranker 0.6B | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 35 ms | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 58 ms | Runs great |
| zerank-1 small | 80Great quality (A) ยท 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking. | 94 ms | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 149 ms | Runs well |
| zerank-2 | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 198 ms | Runs well | 88Excellent quality (A+) ยท 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 342 ms | Runs well |
| LemonDecider 4B | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking. | 115 ms | Runs great | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 200 ms | Runs well |
| Shieldstral 1.0 3B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 190 ms | Runs well | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 328 ms | Runs well |
| Llama Prompt Guard 2 86M | 76Great quality (A) ยท 76 of 100: how capable the model is at F16, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 20 ms | Runs great | 76Great quality (A) ยท 76 of 100: how capable the model is at F16, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 32 ms | Runs great |
| mxbai-rerank-base v2 | 76Great quality (A) ยท 76 of 100: how capable the model is at F16, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 30 ms | Runs great | 76Great quality (A) ยท 76 of 100: how capable the model is at F16, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. | 48 ms | Runs great |
| BGE Reranker v2 Gemma | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 125 ms | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 215 ms | Runs well |
| GLiClass ModernBERT Large v3 | 74Great quality (A) ยท 74 of 100: how capable the model is at F16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 15 ms | Runs great | 74Great quality (A) ยท 74 of 100: how capable the model is at F16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 24 ms | Runs great |
| OpenAI Privacy Filter | 78Great quality (A) ยท 78 of 100: how capable the model is at F16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 5.4 ms | Runs great | 78Great quality (A) ยท 78 of 100: how capable the model is at F16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 6.5 ms | Runs great |
| Qwen3 Reranker 8B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | 399 ms | Runs well | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 692 ms | Slow |
| BGE Reranker v2 M3 | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 36 ms | Runs great | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 61 ms | Runs great |
| DeBERTa v3 Large Zero-shot v2.0 | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 16 ms | Runs great | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 26 ms | Runs great |
| DuoGuard 0.5B | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 31 ms | Runs great | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 53 ms | Runs great |
| Skywork Reward V2 Qwen3 0.6B | 72Great quality (A) ยท 72 of 100: how capable the model is at F16, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | 71 ms | Runs great | 72Great quality (A) ยท 72 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 108 ms | Runs great |
| GTE Reranker ModernBERT Base | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 12 ms | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 19 ms | Runs great |
| Jina Reranker v2 Base Multilingual | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 20 ms | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 32 ms | Runs great |
| ModernBERT Large Zero-shot v2.0 | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 15 ms | Runs great | 70Great quality (A) ยท 70 of 100: how capable the model is at F16, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 24 ms | Runs great |
| Skywork Reward V2 Llama 3.1 8B | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. | 736 ms | Slow | 90Excellent quality (A+) ยท 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 1.3 s | Slow |
| GLiGuard 300M | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 16 ms | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 25 ms | Runs great |
| Granite Embedding Reranker English R2 | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 12 ms | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 19 ms | Runs great |
| Laya | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 16 ms | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 25 ms | Runs great |
| Llama Prompt Guard 2 22M | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 8.0 ms | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 11 ms | Runs great |
| Qwen3Guard Gen 0.6B | 74Great quality (A) ยท 74 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 71 ms | Runs great | 74Great quality (A) ยท 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | 84 ms | Runs great |
| GLiClass Base v3 | 66Good quality (B) ยท 66 of 100: how capable the model is at F16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 9.3 ms | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at F16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 13 ms | Runs great |
| ProtectAI DeBERTa v3 Prompt Injection v2 | 66Good quality (B) ยท 66 of 100: how capable the model is at F16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 14 ms | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at F16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 22 ms | Runs great |
| GLiDE 27B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | 640 ms | Slow | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 1.1 s | Slow |
| DeBERTa v3 Base Zero-shot v2.0 | 64Good quality (B) ยท 64 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 9.0 ms | Runs great | 64Good quality (B) ยท 64 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 13 ms | Runs great |
| GLiNER2 Base v1 | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 9.9 ms | Runs great | 68Good quality (B) ยท 68 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 14 ms | Runs great |
| Qwen3Guard Gen 4B | 83Great quality (A) ยท 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 268 ms | Runs well | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 477 ms | Slow |
| Skywork Reward V2 Qwen3 8B | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | 743 ms | Slow | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 1.3 s | Slow |
| HHEM 2.1 Open | 62Good quality (B) ยท 62 of 100: how capable the model is at F16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 16 ms | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at F16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 26 ms | Runs great |
| LettuceDetect Large | 66Good quality (B) ยท 66 of 100: how capable the model is at F16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 49 ms | Runs great | 66Good quality (B) ยท 66 of 100: how capable the model is at F16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 83 ms | Runs great |
| Qwen3Guard Gen 8B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 539 ms | Slow | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 883 ms | Slow |
| Laya Multilingual | 60Good quality (B) ยท 60 of 100: how capable the model is at F16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 13 ms | Runs great | 60Good quality (B) ยท 60 of 100: how capable the model is at F16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | 20 ms | Runs great |
| Llama 3.1 Nemotron Safety Guard 8B v3 | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 528 ms | Slow | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 865 ms | Slow |
| Granite Guardian 4.1 8B | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | 551 ms | Slow | 84Great quality (A) ยท 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 903 ms | Slow |
| GLiNER Multi v2.1 | 62Good quality (B) ยท 62 of 100: how capable the model is at F16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 12 ms | Runs great | 62Good quality (B) ยท 62 of 100: how capable the model is at F16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 18 ms | Runs great |
| NuExtract 2.0 2B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 145 ms | Runs well | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 273 ms | Runs well |
| NuExtract 2.0 4B | 75Great quality (A) ยท 75 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 313 ms | Runs well | 72Great quality (A) ยท 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 366 ms | Runs well |
| BART Large MNLI | 56Good quality (B) ยท 56 of 100: how capable the model is at F16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 16 ms | Runs great | 56Good quality (B) ยท 56 of 100: how capable the model is at F16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | 24 ms | Runs great |
| WildGuard 7B | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 477 ms | Slow | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 782 ms | Slow |
| Granite Guardian 3.3 8B | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 538 ms | Slow | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 880 ms | Slow |
| ShieldGemma 2B | 58Good quality (B) ยท 58 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. | 130 ms | Runs great | 58Good quality (B) ยท 58 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | 223 ms | Runs well |
| ms-marco MiniLM L6 v2 | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 5.3 ms | Runs great | 52Fair quality (C) ยท 52 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | 6.3 ms | Runs great |
| Llama Guard 3 1B | 56Good quality (B) ยท 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | 102 ms | Runs great | 55Good quality (B) ยท 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 154 ms | Runs well |
| ShieldGemma 9B | 64Good quality (B) ยท 64 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | 450 ms | Slow | 64Good quality (B) ยท 64 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 781 ms | Slow |
| Llama Guard 3 8B | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | 528 ms | Slow | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 865 ms | Slow |
| ShieldGemma 27B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 1.3 s | Slow | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 2.3 s | Very slow |
| CompassJudger-2 32B | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 19.7 s | Very slow | 86Excellent quality (A+) ยท 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 28.0 s | Very slow |
| Nemotron Content Safety Reasoning 4B | 77Great quality (A) ยท 77 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | 1.9 s | Very slow | 77Great quality (A) ยท 77 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 2.7 s | Very slow |
| gpt-oss-safeguard 20B | 76Great quality (A) ยท 76 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 1.6 s | Very slow | 76Great quality (A) ยท 76 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 2.1 s | Very slow |
| DynaGuard 8B | 79Great quality (A) ยท 79 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 4.8 s | Very slow | 79Great quality (A) ยท 79 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 6.7 s | Very slow |
| Selene 1 Mini 8B | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 4.8 s | Very slow | 80Great quality (A) ยท 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 6.9 s | Very slow |
| Llama Guard 4 12B | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | 788 ms | Slow | 66Good quality (B) ยท 66 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 1.3 s | Slow |
| CompassJudger-2 7B | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 4.6 s | Very slow | 78Great quality (A) ยท 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 6.5 s | Very slow |
| DynaGuard 1.7B | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | 906 ms | Slow | 69Good quality (B) ยท 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | 1.3 s | Slow |
| Prometheus 2 8x7B | 71Great quality (A) ยท 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | 5.7 s | Very slow | 71Great quality (A) ยท 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | 8.1 s | Very slow |
| GLIDER | 65Good quality (B) ยท 65 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 1.8 s | Very slow | 65Good quality (B) ยท 65 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 2.6 s | Very slow |
| Prometheus 2 7B | 67Good quality (B) ยท 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 3.4 s | Very slow | 68Good quality (B) ยท 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 6.2 s | Very slow |
| Flow Judge v0.1 | 63Good quality (B) ยท 63 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. | 1.8 s | Very slow | 63Good quality (B) ยท 63 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | 2.6 s | Very slow |
| gpt-oss-safeguard 120B | 80Great quality (A) ยท 80 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 10/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run | 80Great quality (A) ยท 80 of 100: how capable the model is at MXFP4, whatever it runs on. Match score 10/100 (quality, fit, speed and memory use) sets the ranking. | โ | Won't run |
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked โMeasuredโ come from a linked benchmark.