Hardware · Apple Silicon

What AI models can the Apple M3 Ultra (256 GB) run?

Unified memory
256 GB
Usable for models
192 GB
Memory bandwidth
819 GB/s
FP16 compute
28.4 TFLOPS
Released
2025

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 256 GB.

Specifications

MakerApple Silicon
TypeApple silicon (two dies joined)
ChipApple M3 Ultra
Unified memory256 GB
Usable for AI models192 GB
Memory bandwidth819 GB/s
FP16 compute28.4 TFLOPS
InterfaceUNIFIED
Released2025

1 product with the Apple M3 Ultra (256 GB)

Best LLMs for the Apple M3 Ultra (256 GB)

Grade · points Model Quant Memory Speed Fit
97 ptsExcellent (S) · 97 of 100 points: Model quality 42.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 Qwen3.8 Flash Next (125B-A6B)Alibaba Q5_K_M 144 GB 59.6 tok/s Runs well
94 ptsExcellent (S) · 94 of 100 points: Model quality 40.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 MiniMax M2.7 (230B-A10B)MiniMax Q4_K_M 158 GB 43.3 tok/s Runs well
94 ptsExcellent (S) · 94 of 100 points: Model quality 39.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 MiniMax M2.5 (230B-A10B)MiniMax Q4_K_M 158 GB 43.3 tok/s Runs well
94 ptsExcellent (S) · 94 of 100 points: Model quality 39.3/45, Memory fit 25.0/25, Speed 14.9/15, Memory use 15.0/15 Step 3.5 Flash (196B-A11B)StepFun Q4_K_M 138 GB 39.4 tok/s Runs well
93 ptsExcellent (S) · 93 of 100 points: Model quality 38.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 MiniMax M2.1 (230B-A10B)MiniMax Q4_K_M 158 GB 43.3 tok/s Runs well
93 ptsExcellent (S) · 93 of 100 points: Model quality 38.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 MiniMax M2 (230B-A10B)MiniMax Q4_K_M 158 GB 43.3 tok/s Runs well
91 ptsExcellent (S) · 91 of 100 points: Model quality 40.2/45, Memory fit 25.0/25, Speed 14.2/15, Memory use 12.6/15 Qwen3.5 122B-A10BAlibaba Q5_K_M 105 GB 35.8 tok/s Runs well
91 ptsExcellent (S) · 91 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 Ling 3.0 Flash (127B-A5B)Ant Group (inclusionAI) Q8_0 151 GB 43.2 tok/s Runs well
91 ptsExcellent (S) · 91 of 100 points: Model quality 38.3/45, Memory fit 25.0/25, Speed 14.5/15, Memory use 13.7/15 Laguna S 2.1 (118B-A8B)Poolside Q6_K 114 GB 37.5 tok/s Runs well
91 ptsExcellent (S) · 91 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.3/15 Qwen3-Next 80B-A3B ThinkingAlibaba Q8_0 102 GB 73.5 tok/s Runs well
90 ptsExcellent (S) · 90 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.1/15 Qwen3 Coder Next (80B-A3B)Alibaba Q8_0 101 GB 73.5 tok/s Runs well
90 ptsExcellent (S) · 90 of 100 points: Model quality 36.7/45, Memory fit 24.0/25, Speed 14.5/15, Memory use 15.0/15 MiMo V2.6 Flash (310B-A15B)Xiaomi Q3_K_M 168 GB 37.3 tok/s Tight fit

Best Image models for the Apple M3 Ultra (256 GB)

Grade · points Model Precision Memory Time per output Fit
84 ptsGreat (A) · 84 of 100 points: Model quality 41.4/45, Memory fit 25.0/25, Speed 3.8/15, Memory use 13.9/15 FLUX.2 [dev]Black Forest Labs FP16 116 GB 2.5 min Runs well
82 ptsGreat (A) · 82 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15 FLUX.2 [klein] 9BBlack Forest Labs FP16 37.1 GB 7.6 s Runs well
80 ptsGreat (A) · 80 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 2.7/15 Z-Image TurboAlibaba (Tongyi) FP16 22.7 GB 11 s Runs well
78 ptsGreat (A) · 78 of 100 points: Model quality 39.7/45, Memory fit 25.0/25, Speed 3.1/15, Memory use 10.4/15 HunyuanImage 3.0 (80B-A13B)Tencent FP8 86.9 GB 3.8 min Runs well
77 ptsGreat (A) · 77 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.2/15 FLUX.2 [klein] 4BBlack Forest Labs FP16 18.1 GB 4.2 s Runs well
75 ptsGreat (A) · 75 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15 FLUX.1 [schnell]Black Forest Labs FP16 36.4 GB 8.5 s Runs well

Best Video models for the Apple M3 Ultra (256 GB)

Grade · points Model Precision Memory Time per output Fit
77 ptsGreat (A) · 77 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 4.9/15, Memory use 9.1/15 LTX-2.3Lightricks FP16 75.4 GB 28.2 min Runs well
75 ptsGreat (A) · 75 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 4.9/15, Memory use 9.1/15 LTX-2 (19B, video + audio)Lightricks FP16 75.4 GB 28.2 min Runs well
73 ptsGreat (A) · 73 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 2.4/15, Memory use 7.0/15 Wan 2.2 I2V A14BAlibaba FP16 58.0 GB 1.9 h Runs well
73 ptsGreat (A) · 73 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 2.4/15, Memory use 7.0/15 Wan 2.2 T2V A14BAlibaba FP16 58.0 GB 1.9 h Runs well
71 ptsGood (B) · 71 of 100 points: Model quality 30.0/45, Memory fit 25.0/25, Speed 13.3/15, Memory use 3.6/15 LTX-Video 13B 0.9.8 (distilled)Lightricks FP8 30.1 GB 3.8 min Runs well

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.