Apple Mac Studio (2025) (M4 Max or M3 Ultra)

Hardware · Apple Silicon

What AI models can the Apple M3 Ultra (512 GB) run?

Unified memory
512 GB
Usable for models
384 GB
Memory bandwidth
819 GB/s
FP16 compute
28.4 TFLOPS
Released
2025
615 / 621 LLMs run Best: MiMo V2.6 Flash (310B-A15B) 61 / 61 Image models run Best: FLUX.2 [klein] 9B 43 / 43 Video models run Best: LTX-2.5 (22B, video + audio) 70 / 70 Decision models run Best: Jina Reranker v3.5

macOS lets the GPU use roughly two-thirds to three-quarters of unified memory by default, so we plan around that figure rather than the full 512 GB.

More about this hardware

MakerApple Silicon
TypeApple silicon (two dies joined)
ChipApple M3 Ultra

Apps that run models well on the Apple M3 Ultra (512 GB)

macOS

  1. LM StudioRuns the MLX build, typically 20–35% faster than GGUF for a model with 8B active parameters on Apple silicon.
  2. OllamaOne command with Ollama’s own build of this model.
  3. MLX LMRuns the MLX build, typically 20–35% faster than GGUF for a model with 8B active parameters on Apple silicon.
  4. JanRuns the MLX build, typically 20–35% faster than GGUF for a model with 8B active parameters on Apple silicon.

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

1 product with the Apple M3 Ultra (512 GB)

Best LLMs for the Apple M3 Ultra (512 GB)

Quality Model Quant Memory Speed Runs
93Excellent quality (A+) · 93 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2.6 Flash (310B-A15B)Xiaomi Q5_K_M 252 GB 40.9 tok/s Runs great
93Excellent quality (A+) · 93 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. DeepSeek V4 Flash (284B-A13B)DeepSeek Q6_K 281 GB 39.6 tok/s Runs great
96Excellent quality (A+) · 96 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. Qwen3.8 Flash Next (125B-A6B)Alibaba Q8_0 222 GB 57.4 tok/s Runs great
92Excellent quality (A+) · 92 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. Inkling Small (276B-A12B)Thinking Machines Lab Q6_K 250 GB 42.4 tok/s Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. Naive N0.5 Flash (309B-A15.5B)Naive AI Q5_K_M 251 GB 39.7 tok/s Runs great
91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2.5 (310B-A15B)Xiaomi Q5_K_M 252 GB 40.9 tok/s Runs great
92Excellent quality (A+) · 92 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. Intern-S2 (397B-A17B)Shanghai AI Lab Q5_K_M 318 GB 36.7 tok/s Runs great
92Excellent quality (A+) · 92 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. Nex N2 Pro (397B-A17B)Nex-AGI Q5_K_M 313 GB 36.7 tok/s Runs great
91Excellent quality (A+) · 91 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. DeepSeek V4 Flash (April 2026)DeepSeek Q6_K 270 GB 39.6 tok/s Runs great
92Excellent quality (A+) · 92 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 397B-A17BAlibaba Q5_K_M 318 GB 36.7 tok/s Runs great
91Excellent quality (A+) · 91 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 397B-A17BDeepReinforce Q5_K_M 318 GB 36.7 tok/s Runs great
92Excellent quality (A+) · 92 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. MiniMax M3.1 Flash (428B-A23B)MiniMax Q4_K_M 294 GB 33.4 tok/s Runs great

Best Image models for the Apple M3 Ultra (512 GB)

Quality Model Quant Memory Speed Runs
84Great quality (A) · 84 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. FLUX.2 [klein] 9BBlack Forest Labs FP16 37.1 GB 7.6 s Runs great
92Excellent quality (A+) · 92 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. Krea 2 TurboKrea FP16 37.4 GB 18 s Runs well
84Great quality (A) · 84 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. Z-Image TurboAlibaba (Tongyi) FP16 22.7 GB 11 s Runs great
90Excellent quality (A+) · 90 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. HunyuanImage 3.0 (80B-A13B)Tencent FP16 170 GB 4.2 min Very slow
92Excellent quality (A+) · 92 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. Cosmos3-Super Text2Image (64B)NVIDIA FP16 133 GB 2.8 min Very slow
91Excellent quality (A+) · 91 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. MiniMax H3 ImageMiniMax FP8 72.0 GB 38 s Runs well

Best Video models for the Apple M3 Ultra (512 GB)

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 79/100 (quality, fit, speed and memory use) sets the ranking. LTX-2.5 (22B, video + audio)Lightricks FP16 82.5 GB 6.4 min Runs well
90Excellent quality (A+) · 90 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. MiniMax H3 (33B, video + audio)MiniMax FP16 113 GB 73.3 min Very slow
88Excellent quality (A+) · 88 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. Cosmos3-Super Image2Video (64B)NVIDIA FP16 139 GB 2.6 h Very slow
86Excellent quality (A+) · 86 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. LTX-2.3Lightricks FP16 75.4 GB 28.2 min Slow
88Excellent quality (A+) · 88 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. Kandinsky 6.0 Pro (29B, video + audio)Sber AI FP16 85.4 GB 3.5 h Very slow

Best Decision models for the Apple M3 Ultra (512 GB)

Quality Model Quant Memory Speed Runs
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 32.8 GB 24 ms Runs great
90Excellent quality (A+) · 90 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. Qwen3 Reranker 4BAlibaba Q8_0 35.9 GB 130 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. Intern-Decision 4BShanghai AI Lab F16 40.5 GB 92 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 32.8 GB 24 ms Runs great
88Excellent quality (A+) · 88 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. zerank-2ZeroEntropy Q8_0 35.9 GB 130 ms Runs great
86Excellent quality (A+) · 86 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. Shieldstral 1.0 3BMistral AI Q8_0 35.6 GB 125 ms Runs great
82Great quality (A) · 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. d1 3BLiquid AI F16 37.6 GB 65 ms Runs great
82Great quality (A) · 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. mxbai-rerank-large v2Mixedbread F16 34.5 GB 56 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?