Best models for this Mac

30 modelsCatalog current through February 27, 2026

Mac Pro M2 Ultra 192GB ranked for coding with a most capable bias, using the best available runtime evidence.

Mac

Capability

Bias

Runtime

Use the strongest current runtime evidence for each row.M5 Max watch: 8B 61.6 tok/s · 14B 34.3 tok/s on the current 36GB public anchor.

Rank	Model	Score	Quant	Tok/s	Runtime	Evidence	Headroom	Why it ranks here
1	Qwen 3 32B	274	8bit	22.0 tok/s	Ollama	Estimated	159.0 GB	8bit is the highest practical quality here. 22.0 tok/s estimated from nearby benchmark coverage, with Ollama wrapper on llama.cpp as the best runtime hint. 159.0 GB headroom leaves workable context margin.
2	Qwen3.5-27B	264	8bit	20.6 tok/s	MLX	Estimated	164.4 GB	8bit is the highest practical quality here. 20.6 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 164.4 GB headroom leaves workable context margin.
3	Devstral Small 1.1	262	8bit	43.0 tok/s	LM Studio	Estimated	167.9 GB	8bit is the highest practical quality here. 43.0 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 167.9 GB headroom leaves workable context margin.
4	Devstral Small 2 24B	262	8bit	25.3 tok/s	MLX	Estimated	167.9 GB	8bit is the highest practical quality here. 25.3 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 167.9 GB headroom leaves workable context margin.
5	Llama 3.3 70B	261	8bit	11.8 tok/s	LM Studio	Estimated	123.2 GB	8bit is the highest practical quality here. 11.8 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 123.2 GB headroom leaves workable context margin.
6	Gemma 3 27B	259	8bit	14.5 tok/s	LM Studio	Estimated	162.1 GB	8bit is the highest practical quality here. 14.5 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 162.1 GB headroom leaves workable context margin.
7	DeepSeek R1 Distill Qwen 32B	252	8bit	Measure it	Best available	Fit-first	159.0 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 159.0 GB headroom leaves workable context margin.
8	Mistral Small 3.1 24B	240	8bit	Measure it	Best available	Fit-first	167.9 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 167.9 GB headroom leaves workable context margin.
9	Magistral Small	240	8bit	Measure it	Best available	Fit-first	167.9 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 167.9 GB headroom leaves workable context margin.
10	Nemotron-3-Nano-30B-A3B	233	8bit	43.7 tok/s	llama.cpp	Estimated	163.2 GB	8bit is the highest practical quality here. 43.7 tok/s estimated from nearby benchmark coverage, with llama.cpp backend as the best runtime hint. 163.2 GB headroom leaves workable context margin.
11	Qwen 3 30B-A3B	233	8bit	76.7 tok/s	MLX	Estimated	162.3 GB	8bit is the highest practical quality here. 76.7 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 162.3 GB headroom leaves workable context margin.
12	Qwen3-Coder-30B-A3B	233	8bit	58.5 tok/s	llama.cpp	Estimated	162.7 GB	8bit is the highest practical quality here. 58.5 tok/s estimated from nearby benchmark coverage, with llama.cpp backend as the best runtime hint. 162.7 GB headroom leaves workable context margin.
13	Qwen3.5-35B-A3B	232	8bit	80.0 tok/s	MLX	Estimated	158.3 GB	8bit is the highest practical quality here. 80.0 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 158.3 GB headroom leaves workable context margin.
14	GLM-4.7-Flash	232	8bit	36.8 tok/s	llama.cpp	Estimated	156.2 GB	8bit is the highest practical quality here. 36.8 tok/s estimated from nearby benchmark coverage, with llama.cpp backend as the best runtime hint. 156.2 GB headroom leaves workable context margin.
15	Qwen 3 8B	211	8bit	63.1 tok/s	LM Studio	Estimated	182.7 GB	8bit is the highest practical quality here. 63.1 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 182.7 GB headroom leaves workable context margin.
16	Llama 2 7B	209	8bit	94.3 tok/s	llama.cpp	Estimated	181.2 GB	8bit is the highest practical quality here. 94.3 tok/s estimated from nearby benchmark coverage, with llama.cpp backend as the best runtime hint. 181.2 GB headroom leaves workable context margin.
17	Qwen3.5-9B	206	8bit	15.0 tok/s	llama.cpp	Estimated	182.1 GB	8bit is the highest practical quality here. 15.0 tok/s estimated from nearby benchmark coverage, with llama.cpp backend as the best runtime hint. 182.1 GB headroom leaves workable context margin.
18	Qwen 2.5 14B	199	8bit	Measure it	Best available	Fit-first	177.0 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 177.0 GB headroom leaves workable context margin.
19	Phi-4 14B	198	8bit	Measure it	Best available	Fit-first	177.5 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 177.5 GB headroom leaves workable context margin.
20	Llama 3.1 8B	189	8bit	Measure it	Best available	Fit-first	183.0 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 183.0 GB headroom leaves workable context margin.
21	GLM-4.5-Air	189	8bit	54.0 tok/s	MLX	Estimated	91.3 GB	8bit is the highest practical quality here. 54.0 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 91.3 GB headroom leaves workable context margin.
22	Qwen 2.5 7B	189	8bit	Measure it	Best available	Fit-first	184.0 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 184.0 GB headroom leaves workable context margin.
23	Mistral 7B v0.3	188	8bit	Measure it	Best available	Fit-first	183.7 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 183.7 GB headroom leaves workable context margin.
24	Qwen3.5-122B-A10B	186	8bit	57.0 tok/s	MLX	Estimated	77.1 GB	8bit is the highest practical quality here. 57.0 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 77.1 GB headroom leaves workable context margin.
25	Qwen3-Coder-Next	176	8bit	74.0 tok/s	MLX	Estimated	116.2 GB	8bit is the highest practical quality here. 74.0 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 116.2 GB headroom leaves workable context margin.
26	Gemma 3 4B	150	8bit	100.5 tok/s	LM Studio	Estimated	186.4 GB	8bit is the highest practical quality here. 100.5 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 186.4 GB headroom leaves workable context margin.
27	Qwen 3 4B	150	8bit	143.2 tok/s	MLX	Estimated	186.6 GB	8bit is the highest practical quality here. 143.2 tok/s estimated from nearby benchmark coverage, with MLX backend as the best runtime hint. 186.6 GB headroom leaves workable context margin.
28	Qwen 3 0.6B	145	8bit	184.4 tok/s	LM Studio	Estimated	190.8 GB	8bit is the highest practical quality here. 184.4 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 190.8 GB headroom leaves workable context margin.
29	Qwen 3 235B-A22B	141	Q5_K_M	27.4 tok/s	LM Studio	Estimated	32.8 GB	Q5_K_M is the highest practical quality here. 27.4 tok/s estimated from nearby benchmark coverage, with LM Studio wrapper on mixed as the best runtime hint. 32.8 GB headroom leaves workable context margin.
30	Llama 3.2 1B	124	8bit	Measure it	Best available	Fit-first	190.1 GB	8bit is the highest practical quality here. Speed still needs direct measurement. 190.1 GB headroom leaves workable context margin.

Machine

192GBUnified memory

$6,999MSRP

mac_proForm factor

m2-ultraChip