AMD Instinct MI250X

Hardware · AMD

What AI models can the AMD Instinct MI250X run?

VRAM
128 GB
Usable for models
128 GB
Memory bandwidth
3,277 GB/s
FP16 compute
95.7 TFLOPS
Released
2021
558 / 621 LLMs run Best: Nemotron 3.5 Bolt 74B-A9B 61 / 61 Image models run Best: MiniMax H3 Image 43 / 43 Video models run Best: LTX-2.5 (22B, video + audio) 70 / 70 Decision models run Best: Qwen3 Reranker 4B

More about this hardware

MakerAMD
TypeDatacenter GPU module
InterfacePCIe 4.0

Apps that run models well on the AMD Instinct MI250X

Linux

  1. vLLMThe full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.
  2. LemonadeAMD’s own app: llama.cpp tuned for Radeon on Vulkan and ROCm.
  3. LM StudioPoint-and-click: search, download and chat in one window.
  4. OllamaOne command with Ollama’s own build of this model.

Windows

  1. LemonadeAMD’s own app: llama.cpp tuned for Radeon on Vulkan and ROCm.
  2. LM StudioPoint-and-click: search, download and chat in one window.
  3. OllamaOne command with Ollama’s own build of this model.
  4. llama.cppUsually the first to support new model architectures.

For a typical model that fits in memory. Each model page picks the best app for that model on this hardware. All apps

Best LLMs for the AMD Instinct MI250X

Quality Model Quant Memory Speed Runs
92Excellent quality (A+) · 92 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. Nemotron 3.5 Bolt 74B-A9BNVIDIA Q8_0 86.5 GB 112 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 35B-A3BAlibaba F16 79.1 GB 124 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 35B-A3BAlibaba F16 79.1 GB 124 tok/s Runs great
93Excellent quality (A+) · 93 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. MiniCPM-V 4.7 35B-A3BOpenBMB F16 77.6 GB 124 tok/s Runs great
91Excellent quality (A+) · 91 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. Kolibri 1 (78B-A3.5B)Aleph Alpha Q8_0 90.0 GB 173 tok/s Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Nemotron 3.5 Super VL 120B-A12BNVIDIA Q5_K_M 95.4 GB 130 tok/s Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Ornith 1.5 122B-A10BDeepReinforce Q5_K_M 96.0 GB 143 tok/s Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Flint (40B-A2.5B)Recursive F16 86.8 GB 134 tok/s Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Nex N2 Mini (35B-A3B)Nex-AGI F16 77.4 GB 124 tok/s Runs great
91Excellent quality (A+) · 91 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.8 Flash Next (125B-A6B)Alibaba Q4_K_M 117 GB 196 tok/s Runs welltight fit
91Excellent quality (A+) · 91 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Sage-A1.5 35B-A3BInternScience F16 79.1 GB 124 tok/s Runs great
89Excellent quality (A+) · 89 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 122B-A10BAlibaba Q5_K_M 96.0 GB 143 tok/s Runs great

Best Image models for the AMD Instinct MI250X

Quality Model Quant Memory Speed Runs
91Excellent quality (A+) · 91 of 100: how capable the model is at FP8, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking. MiniMax H3 ImageMiniMax FP8 72.0 GB 12 s Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at FP16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. Krea 2 TurboKrea FP16 37.4 GB 5.8 s Runs great
89Excellent quality (A+) · 89 of 100: how capable the model is at FP16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Qwen-Image 2.1 TurboAlibaba FP16 46.7 GB 16 s Runs well
86Excellent quality (A+) · 86 of 100: how capable the model is at FP16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Ming-Image 0.1 DesignAnt Group (inclusionAI) FP16 63.9 GB 22 s Runs well
88Excellent quality (A+) · 88 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. HunyuanImage 3.0 (80B-A13B)Tencent FP8 86.9 GB 73 s Slow
91Excellent quality (A+) · 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. FLUX 3 Image KleinBlack Forest Labs FP16 22.4 GB 7.7 s Runs great

Best Video models for the AMD Instinct MI250X

Quality Model Quant Memory Speed Runs
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. LTX-2.5 (22B, video + audio)Lightricks FP16 82.5 GB 2.0 min Runs great
86Excellent quality (A+) · 86 of 100: how capable the model is at FP16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. LTX-2.3Lightricks FP16 75.4 GB 9.0 min Runs well
82Great quality (A) · 82 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. LTX-2 (19B, video + audio)Lightricks FP16 75.4 GB 9.0 min Runs well
88Excellent quality (A+) · 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. Kandinsky 6.0 Pro (29B, video + audio)Sber AI FP16 85.4 GB 65.8 min Very slow
88Excellent quality (A+) · 88 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. MiniMax H3 (33B, video + audio)MiniMax FP8 66.8 GB 21.0 min Slow

Best Decision models for the AMD Instinct MI250X

Quality Model Quant Memory Speed Runs
90Excellent quality (A+) · 90 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Qwen3 Reranker 4BAlibaba F16 15.3 GB 44 ms Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at F16, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking. Qwen3 Reranker 8BAlibaba F16 23.7 GB 85 ms Runs great
88Excellent quality (A+) · 88 of 100: how capable the model is at F16, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking. zerank-2ZeroEntropy F16 15.3 GB 44 ms Runs great
86Excellent quality (A+) · 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. Jina Reranker v3.5Jina AI F16 8.4 GB 9.9 ms Runs great
86Excellent quality (A+) · 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. Shieldstral 1.0 3BMistral AI F16 14.9 GB 42 ms Runs great
90Excellent quality (A+) · 90 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. Skywork Reward V2 Llama 3.1 8BSkywork Q8_0 10.4 GB 145 ms Runs well
84Great quality (A) · 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. Intern-Decision 4BShanghai AI Lab F16 16.1 GB 29 ms Runs great
84Great quality (A) · 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. Jina Reranker v3Jina AI F16 8.4 GB 9.9 ms Runs great

Best LLMs by task on this hardware Open in Model Finder

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 “Measured” come from a linked benchmark.

Was this page helpful?