Your rig
M3 Ultra · 256 GB · 256 GB
256 GB unified · 800 GB/s · 192.0 GB addressable by the GPU
Which one should I actually run?
Best models for your M3 Ultra · 256 GB
What do you want to do? Each row is the best-ranked model on that use case's shortlist that fits at Q3_K_M and 8K. The ordering is editorial; the sizes and speeds are computed. How the shortlists work.
All
General assistantCodingReasoning & mathsWritingVisionAgents & tool useTranslationLow latencyLong context
| Recommendation | Model | Why | Quant · total | Tok/s est. | |
|---|---|---|---|---|---|
| Best overall | DeepSeek V4 Flash 284B-A13B New DeepSeek · MIT |
Frontier-adjacent at 13B active; the largest model a 128–192 GB machine runs at a useful speed. | Q3_K_M · 130.3 GB | 35 | Runs great |
| Best for coding | DeepSeek V4 Flash 284B-A13B New DeepSeek · MIT |
Frontier coding at 13B active for 128 GB+ machines. | Q3_K_M · 130.3 GB | 35 | Runs great |
| Best reasoning | DeepSeek V4 Flash 284B-A13B New DeepSeek · MIT |
Most of V4 Pro's reasoning at 13B active. | Q3_K_M · 130.3 GB | 35 | Runs great |
| Best for writing | Llama 3.3 70B Instruct Meta · Llama 3.3 Community |
The creative-writing favourite: consistent voice, takes direction, handles fiction framing. | Q3_K_M · 35.2 GB | 14 | Runs great |
| Best vision | Qwen3.8 27B New this week Alibaba · Apache 2.0 |
Image and video in, OSWorld 84 — the strongest local vision-language model at 24 GB. | Q3_K_M · 13.8 GB | 35 | Runs great |
| Best for agents | DeepSeek V4 Flash 284B-A13B New DeepSeek · MIT |
Long agent trajectories at 13B active. | Q3_K_M · 130.3 GB | 35 | Runs great |
| Best translation | Qwen3.8 27B New this week Alibaba · Apache 2.0 |
119 languages, and Qwen translations read as idiomatic rather than literal. | Q3_K_M · 13.8 GB | 35 | Runs great |
| Fastest good model | Qwen3.6 35B-A3B Alibaba · Apache 2.0 |
3.3B active per token: the fastest model that still competes with dense 27Bs. | Q3_K_M · 17.1 GB | 139 | Runs great |
| Best long context | DeepSeek V4 Flash 284B-A13B New DeepSeek · MIT |
A 1M-token window with a compressed latent cache. | Q3_K_M · 135.9 GB at 128K context |
35 | Runs great |
Everything that fits
Runs great
68 of 76 · fits with headroom for a longer prompt| Model | Params | Quant | Weights | +KV 8K | Total | VRAM headroom | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash 284B-A13B New | 284B MoE / 13B act | Q3_K_M | 129.3 GB | 0.38 GB | 130.3 GB |
61.7 GB free
|
35 | MLX | Run |
| Qwen3 235B-A22B | 235B MoE / 22B act | Q3_K_M | 107.0 GB | 1.47 GB | 109.0 GB |
83.0 GB free
|
21 | MLX | Run |
| Qwen3.5 122B-A10B | 125B MoE / 10B act | Q3_K_M | 56.9 GB | 0.19 GB | 57.7 GB |
134.3 GB free
|
46 | MLX | Run |
| Nemotron 3 Super 120B-A12B | 124B MoE / 12B act | Q3_K_M | 56.4 GB | 0.06 GB | 57.1 GB |
134.9 GB free
|
38 | MLX | Run |
| Ling 3.0 Flash 124B-A5B New | 124B MoE / 5.1B act | Q3_K_M | 56.4 GB | 0.06 GB | 57.1 GB |
134.9 GB free
|
90 | MLX | Run |
| Mistral Small 4 119B-A6B | 119B MoE / 6.5B act | Q3_K_M | 54.2 GB | 0.18 GB | 54.9 GB |
137.1 GB free
|
71 | MLX | Run |
| gpt-oss 120B | 117B MoE / 5.1B act | MXFP4 | 60.5 GB | 0.29 GB | 61.4 GB |
130.6 GB free
|
79 | MLX | Run |
| Llama 4 Scout 109B-A17B | 109B MoE / 17B act | Q3_K_M | 49.6 GB | 1.50 GB | 51.7 GB |
140.3 GB free
|
27 | MLX | Run |
| Llama 3.3 70B Instruct | 70.6B | Q3_K_M | 32.1 GB | 2.50 GB | 35.2 GB |
156.8 GB free
|
14 | MLX | Run |
| DeepSeek-R1-Distill-Llama 70B | 70.6B | Q3_K_M | 32.1 GB | 2.50 GB | 35.2 GB |
156.8 GB free
|
14 | MLX | Run |
| Qwen3.6 35B-A3B | 35.9B MoE / 3.3B act | Q3_K_M | 16.3 GB | 0.16 GB | 17.1 GB |
174.9 GB free
|
139 | MLX | Run |
| Ornith 1.5 35B-A3B New this week | 35.9B MoE / 3.3B act | Q3_K_M | 16.3 GB | 0.16 GB | 17.1 GB |
174.9 GB free
|
139 | MLX | Run |
| LLM-jp 4 33B Thinking New this week | 33.2B | Q3_K_M | 15.1 GB | 2.00 GB | 17.7 GB |
174.3 GB free
|
30 | MLX | Run |
| Qwen3 32B | 32.8B | Q3_K_M | 14.9 GB | 2.00 GB | 17.5 GB |
174.5 GB free
|
30 | MLX | Run |
| DeepSeek-R1-Distill-Qwen 32B | 32.8B | Q3_K_M | 14.9 GB | 2.00 GB | 17.5 GB |
174.5 GB free
|
30 | MLX | Run |
| Qwen2.5-Coder 32B | 32.8B | Q3_K_M | 14.9 GB | 2.00 GB | 17.5 GB |
174.5 GB free
|
30 | MLX | Run |
| Olmo 3.1 32B Instruct | 32.2B | Q3_K_M | 14.7 GB | 1.25 GB | 16.5 GB |
175.5 GB free
|
30 | MLX | Run |
| Nemotron 3.5 Lightning 30B-A3B New | 31.6B MoE / 3.2B act | Q3_K_M | 14.4 GB | 0.05 GB | 15.0 GB |
177.0 GB free
|
143 | MLX | Run |
| Gemma 4 31B | 31.3B | Q3_K_M | 14.2 GB | 2.03 GB | 16.9 GB |
175.1 GB free
|
31 | MLX | Run |
| GLM-4.7-Flash 30B-A3B | 31.2B MoE / 3B act | Q3_K_M | 14.2 GB | 0.41 GB | 15.2 GB |
176.8 GB free
|
153 | MLX | Run |
| Qwen3 30B-A3B | 30.5B MoE / 3.3B act | Q3_K_M | 13.9 GB | 0.75 GB | 15.2 GB |
176.8 GB free
|
139 | MLX | Run |
| Qwen3 Coder 30B-A3B | 30.5B MoE / 3.3B act | Q3_K_M | 13.9 GB | 0.75 GB | 15.2 GB |
176.8 GB free
|
139 | MLX | Run |
| Muse Glimmer 30B New | 29.8B | Q3_K_M | 13.6 GB | 0.18 GB | 14.3 GB |
177.7 GB free
|
33 | MLX | Run |
| Granite 4.1 30B | 28.9B | Q3_K_M | 13.2 GB | 2.00 GB | 15.8 GB |
176.2 GB free
|
34 | MLX | Run |
| Qwen3.8 27B New this week | 27.8B | Q3_K_M | 12.7 GB | 0.50 GB | 13.8 GB |
178.2 GB free
|
35 | MLX | Run |
| Qwen3.6 27B | 27.8B | Q3_K_M | 12.7 GB | 0.50 GB | 13.8 GB |
178.2 GB free
|
35 | MLX | Run |
| Gemma 3 27B | 27.4B | Q3_K_M | 12.5 GB | 1.03 GB | 14.1 GB |
177.9 GB free
|
36 | MLX | Run |
| Gemma 4 26B-A4B | 26.5B MoE / 3.8B act | Q3_K_M | 12.1 GB | 0.51 GB | 13.2 GB |
178.8 GB free
|
121 | MLX | Run |
| Devstral Small 2 24B | 24B | Q3_K_M | 10.9 GB | 1.25 GB | 12.8 GB |
179.2 GB free
|
41 | MLX | Run |
| Mistral Small 3.2 24B | 23.6B | Q3_K_M | 10.7 GB | 1.25 GB | 12.6 GB |
179.4 GB free
|
42 | MLX | Run |
| gpt-oss 20B | 20.9B MoE / 3.6B act | MXFP4 | 10.8 GB | 0.19 GB | 11.6 GB |
180.4 GB free
|
112 | MLX | Run |
| Qwen3 14B | 14.8B | Q3_K_M | 6.7 GB | 1.25 GB | 8.6 GB |
183.4 GB free
|
66 | MLX | Run |
| DeepSeek-R1-Distill-Qwen 14B | 14.8B | Q3_K_M | 6.7 GB | 1.50 GB | 8.8 GB |
183.2 GB free
|
66 | MLX | Run |
| Qwen2.5-Coder 14B | 14.8B | Q3_K_M | 6.7 GB | 1.50 GB | 8.8 GB |
183.2 GB free
|
66 | MLX | Run |
| Phi-4 14B | 14.7B | Q3_K_M | 6.7 GB | 1.56 GB | 8.9 GB |
183.1 GB free
|
67 | MLX | Run |
| Ministral 3 14B | 13.9B | Q3_K_M | 6.3 GB | 1.25 GB | 8.2 GB |
183.8 GB free
|
71 | MLX | Run |
| Gemma 3 12B | 12.2B | Q3_K_M | 5.6 GB | 0.81 GB | 7.0 GB |
185.0 GB free
|
80 | MLX | Run |
| Mistral NeMo 12B | 12.2B | Q3_K_M | 5.6 GB | 1.25 GB | 7.4 GB |
184.6 GB free
|
80 | MLX | Run |
| Gemma 4 12B | 12B | Q3_K_M | 5.5 GB | 0.81 GB | 6.9 GB |
185.1 GB free
|
82 | MLX | Run |
| Qwen3.5 9B | 9.65B | Q3_K_M | 4.4 GB | 0.25 GB | 5.2 GB |
186.8 GB free
|
102 | MLX | Run |
| Ornith 1.5 9B New this week | 9.41B | Q3_K_M | 4.3 GB | 0.25 GB | 5.1 GB |
186.9 GB free
|
104 | MLX | Run |
| Ministral 3 8B | 8.92B | Q3_K_M | 4.1 GB | 1.06 GB | 5.7 GB |
186.3 GB free
|
110 | MLX | Run |
| Granite 4.1 8B | 8.79B | Q3_K_M | 4.0 GB | 1.25 GB | 5.9 GB |
186.1 GB free
|
112 | MLX | Run |
| LFM2.5 8B-A1B | 8.47B MoE / 1.5B act | Q3_K_M | 3.9 GB | 0.09 GB | 4.5 GB |
187.5 GB free
|
306 | MLX | Run |
| Fara 7B | 8.29B | Q3_K_M | 3.8 GB | 0.44 GB | 4.8 GB |
187.2 GB free
|
118 | MLX | Run |
| Qwen3 8B | 8.19B | Q3_K_M | 3.7 GB | 1.13 GB | 5.5 GB |
186.5 GB free
|
120 | MLX | Run |
| Llama 3.1 8B Instruct | 8.03B | Q3_K_M | 3.7 GB | 1.00 GB | 5.3 GB |
186.7 GB free
|
122 | MLX | Run |
| Gemma 4 E4B | 8.0B | Q3_K_M | 3.6 GB | 0.14 GB | 4.4 GB |
187.6 GB free
|
123 | MLX | Run |
| Ling 3.0 Tiny 7.9B-A1.3B New | 7.9B MoE / 1.3B act | Q3_K_M | 3.6 GB | 0.05 GB | 4.2 GB |
187.8 GB free
|
353 | MLX | Run |
| DeepSeek-R1-Distill-Qwen 7B | 7.62B | Q3_K_M | 3.5 GB | 0.44 GB | 4.5 GB |
187.5 GB free
|
129 | MLX | Run |
| Qwen2.5-Coder 7B | 7.62B | Q3_K_M | 3.5 GB | 0.44 GB | 4.5 GB |
187.5 GB free
|
129 | MLX | Run |
| Olmo 3 7B Instruct | 7.3B | Q3_K_M | 3.3 GB | 2.50 GB | 6.4 GB |
185.6 GB free
|
135 | MLX | Run |
| Mistral 7B Instruct v0.3 | 7.25B | Q3_K_M | 3.3 GB | 1.00 GB | 4.9 GB |
187.1 GB free
|
135 | MLX | Run |
| Gemma 4 E2B | 5.1B | Q3_K_M | 2.3 GB | 0.07 GB | 3.0 GB |
189.0 GB free
|
193 | MLX | Run |
| Qwen3.5 4B | 4.66B | Q3_K_M | 2.1 GB | 0.25 GB | 3.0 GB |
189.0 GB free
|
211 | MLX | Run |
| Gemma 3 4B | 4.3B | Q3_K_M | 2.0 GB | 0.30 GB | 2.9 GB |
189.1 GB free
|
228 | MLX | Run |
| Qwen3 4B | 4.02B | Q3_K_M | 1.8 GB | 1.13 GB | 3.6 GB |
188.4 GB free
|
244 | MLX | Run |
| Ministral 3 3B | 3.85B | Q3_K_M | 1.8 GB | 0.81 GB | 3.2 GB |
188.8 GB free
|
255 | MLX | Run |
| Phi-4-mini 3.8B | 3.84B | Q3_K_M | 1.7 GB | 1.00 GB | 3.3 GB |
188.7 GB free
|
256 | MLX | Run |
| Granite 4.1 3B | 3.4B | Q3_K_M | 1.5 GB | 0.63 GB | 2.8 GB |
189.2 GB free
|
289 | MLX | Run |
| Llama 3.2 3B Instruct | 3.21B | Q3_K_M | 1.5 GB | 0.88 GB | 2.9 GB |
189.1 GB free
|
306 | MLX | Run |
| LFM2.5 2.6B New | 2.7B | Q3_K_M | 1.2 GB | 0.13 GB | 2.0 GB |
190.0 GB free
|
364 | MLX | Run |
| Qwen3.5 2B | 2.27B | Q3_K_M | 1.0 GB | 0.09 GB | 1.7 GB |
190.3 GB free
|
433 | MLX | Run |
| Qwen3 1.7B | 1.72B | Q3_K_M | 0.8 GB | 0.88 GB | 2.3 GB |
189.7 GB free
|
571 | MLX | Run |
| Llama 3.2 1B Instruct | 1.24B | Q3_K_M | 0.6 GB | 0.25 GB | 1.4 GB |
190.6 GB free
|
792 | MLX | Run |
| Gemma 3 1B | 1.0B | Q3_K_M | 0.5 GB | 0.04 GB | 1.1 GB |
190.9 GB free
|
982 | MLX | Run |
| Qwen3.5 0.8B | 0.87B | Q3_K_M | 0.4 GB | 0.09 GB | 1.1 GB |
190.9 GB free
|
1129 | MLX | Run |
| Qwen3 0.6B | 0.6B | Q3_K_M | 0.3 GB | 0.88 GB | 1.7 GB |
190.3 GB free
|
1637 | MLX | Run |
Runs — tight
2 models · fits at 8K, drop context or quant to go longer| Model | Params | Quant | Weights | +KV 8K | Total | VRAM headroom | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| Qwen3.5 397B-A17B | 403B MoE / 17B act | Q3_K_M | 183.4 GB | 0.23 GB | 184.3 GB |
7.7 GB free
|
27 | MLX | Run |
| Ornith 1.5 397B-A17B New this week | 397B MoE / 17B act | Q3_K_M | 180.7 GB | 0.23 GB | 181.5 GB |
10.5 GB free
|
27 | MLX | Run |
Won't fit in VRAM
6 models · some are possible with CPU offload| Model | Params | Quant | Weights | +KV 8K | Total | Over budget | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| Kimi K3 2.8T-A104B New | 2780B MoE / 104B act | Q3_K_M | 1265.4 GB | 0.21 GB | 1266.2 GB | +1074.2 GB | ~4.4 | MLX | Why |
| Qwen3.8 2.4T-A95B New | 2446B MoE / 95B act | Q3_K_M | 1113.4 GB | 0.72 GB | 1114.7 GB | +922.7 GB | ~4.8 | MLX | Why |
| DeepSeek V4 Pro 1.6T-A49B New | 1650B MoE / 49B act | Q3_K_M | 751.1 GB | 0.54 GB | 752.2 GB | +560.2 GB | ~9.4 | MLX | Why |
| Kimi K2.6 1T-A32B | 1027B MoE / 32B act | Q3_K_M | 467.5 GB | 0.54 GB | 468.6 GB | +276.6 GB | ~14 | MLX | Why |
| GLM-5.2 744B-A40B | 753B MoE / 40B act | Q3_K_M | 342.8 GB | 0.69 GB | 344.0 GB | +152.0 GB | ~11 | MLX | Why |
| DeepSeek-R1 671B | 671B MoE / 37B act | Q3_K_M | 305.4 GB | 0.54 GB | 306.6 GB | +114.6 GB | ~12 | MLX | Why |
Totals = quantised weights + f16 KV cache at 8K tokens + 0.6 GB runtime overhead, against 192.0 GB usable unified memory. Tokens per second are estimated from 800 GB/s peak memory bandwidth at batch 1 — not measured. Read the method.