Macs

Which AI models run on the M2 Pro (32 GB)?

The M2 Pro (32 GB) has 32 GB of memory, about 21 GB of it usable for a model, and 200 GB/s of memory bandwidth. 70 of the 110 models in our catalog run on it at 4-bit or better.

Data as of 2 November 2026 · Vendor specs

Compare with other machines

32 GBmemory ~21 GB usable for a model

200 GB/sbandwidth, sets the speed

70models run at 4-bit or better

Our pick for this machine Qwen3-Coder-30B-A3B Q4_K_M · 20 of 21 GB · ~45 tok/s ModelFit est. The largest model on our shortlist of current models that runs at 4-bit or better and 10 tok/s or more.

Models that run on it

Model	Best build	Memory	Speed
Qwen3-Coder-30B-A3B30.5B MoE	Q4_K_M	20 GB	~45 tok/s ModelFit est.
Nemotron 3 Nano 30B-A3B30B MoE	Q4_K_M	19.6 GB	~60 tok/s our est.
Qwen3.8-27B27.3B	Q4_K_M	17.9 GB	~10 tok/s ModelFit est.
Gemma 4 26B-A4B25.2B MoE	Q4_K_M	17.4 GB	~39 tok/s ModelFit est.
gpt-oss-20b21B MoE	MXFP4	15.1 GB	~39 tok/s ModelFit est.
Nemotron Nano 9B v29B	Q8_0	10.8 GB	~12 tok/s our est.
Qwen3.5 9B Instruct9B	Q8_0	12.0 GB	~19 tok/s ModelFit est.
Qwen2.5 7B Instruct7B	Q4_K_M	6.6 GB	~35 tok/s ModelFit est.
Gemma 3 4B Instruct4B	Q8_0	5.0 GB	~36 tok/s ModelFit est.
Olmo 3.1 32B Instruct32B	Q4_K_M	21 GB	~8.2 tok/s slow ModelFit est.
Cohere North Mini Code30B	Q4_K_M	20 GB	no estimate
Nemotron 3.5 Lightning 30B-A3B30B MoE	IQ4_XS	20 GB	~58 tok/s our est.
GLM-4.7-Flash 30B-A3B30B MoE	Q4_K_M	20 GB	~42 tok/s ModelFit est.
Muse Glimmer 30B29.8B	Q4_K_M	18.3 GB	~10 tok/s ModelFit est.
Granite 4.2 30B29.3B	Q4_K_M	17.9 GB	~8.9 tok/s slow ModelFit est.
AREX-2 27B27.4B	Q4_K_M	18.6 GB	~10 tok/s ModelFit est.
Qwen3.5 27B Instruct27B	Q4_K_M	17.4 GB	~10 tok/s ModelFit est.
Qwen3.6 27B27B	Q4_K_M	19.4 GB	~10 tok/s ModelFit est.
Jebadiah 27B27B	Q4_K_M	18.2 GB	~10 tok/s ModelFit est.
Mistral Small 3.124B	Q4_K_M	16.4 GB	~7.3 tok/s slow our est.
LFM2 24B-A2B Instruct24B MoE	Q4_K_M	15.3 GB	~67 tok/s ModelFit est.
Devstral Small 2 24B24B	Q4_K_M	16.6 GB	~11 tok/s ModelFit est.
Mistral Small 22B22B	Q4_K_M	18.4 GB	~12 tok/s ModelFit est.
Qwen2.5 14B Instruct14B	Q4_K_M	12.3 GB	~18 tok/s ModelFit est.
Phi-3 Medium 14B14B	Q4_K_M	12.3 GB	~10 tok/s our est.
Phi-4 14B14B	Q8_0	15.9 GB	~10 tok/s ModelFit est.
Qwen2.5 Coder 14B14B	Q4_K_M	12.3 GB	~18 tok/s ModelFit est.
Qwen3 14B14B	Q8_0	17.3 GB	~11 tok/s ModelFit est.
DeepSeek-R1 Distill Qwen 14B14B	Q8_0	16.0 GB	~11 tok/s ModelFit est.
Ministral 3 14B13.5B	Q4_K_M	10.3 GB	~18 tok/s ModelFit est.
Mellum2.1 12B-A2.5B Thinking12.1B MoE	Q4_K_M	9.3 GB	~66 tok/s our est.
Mistral Nemo 12B12B	Q8_0	13.4 GB	~13 tok/s ModelFit est.
Gemma 3 12B Instruct12B	Q4_K_M	10.7 GB	~21 tok/s ModelFit est.
Gemma 4 12B12B	Q8_0	14.1 GB	~12 tok/s ModelFit est.
GLM-4-9B-Chat9.4B	Q8_0	11.2 GB	~17 tok/s ModelFit est.
Gemma 2 9B Instruct9B	Q4_K_M	8.2 GB	~23 tok/s ModelFit est.
Ornith 1.0 9B9B	Q4_K_M	6.7 GB	~32 tok/s ModelFit est.
MiMo-V2.6-Distill-Qwen-9B9B	Q4_K_M	6.9 GB	~31 tok/s ModelFit est.
Jebadiah 9B9B	Q8_0	11.0 GB	~19 tok/s ModelFit est.
Granite 4.2 8B8.8B	Q4_K_M	6.1 GB	~28 tok/s ModelFit est.
Granite Switch 4.1 8B8.8B	Q4_K_M	7.6 GB	~23 tok/s ModelFit est.
MiniCPM-V 4.5 8B8.7B	Q4_K_M	6.8 GB	~30 tok/s ModelFit est.
Ministral 3 8B8.4B	Q4_K_M	7.1 GB	~26 tok/s ModelFit est.
LFM2.5 8B-A1B8.3B MoE	Q4_K_M	6.6 GB	~88 tok/s ModelFit est.
Llama 3.1 8B Instruct8B	Q8_0	9.2 GB	~19 tok/s ModelFit est.
Qwen3 8B8B	Q8_0	9.3 GB	~19 tok/s ModelFit est.
Granite 4.1 8B Instruct8B	Q4_K_M	6.6 GB	~27 tok/s ModelFit est.
Mistral 7B Instruct7B	Q8_0	8.4 GB	~20 tok/s ModelFit est.
Qwen2.5 Coder 7B7B	Q4_K_M	6.6 GB	~35 tok/s ModelFit est.
DeepSeek-R1 Distill Qwen 7B7B	Q8_0	8.7 GB	~21 tok/s ModelFit est.
Gemma 4 E4B4.5B	Q8_0	8.7 GB	~32 tok/s ModelFit est.
Qwen3.5 4B Instruct4B	Q8_0	5.4 GB	~34 tok/s ModelFit est.
NVIDIA Nemotron 3.5 Bolt E4B4B	Q4_K_M	9.1 GB	~40 tok/s ModelFit est.
Phi-3 Mini 3.8B3.8B	Q4_K_M	4.3 GB	~34 tok/s our est.
Phi-4 Mini 3.8B3.8B	Q8_0	4.9 GB	~34 tok/s ModelFit est.
Granite 4.2 3B3.7B	Q4_K_M	3.2 GB	~62 tok/s ModelFit est.
Ministral 3 3B3.4B	Q4_K_M	4.1 GB	~46 tok/s ModelFit est.
Qwen2.5 3B Instruct3B	Q4_K_M	3.6 GB	~72 tok/s ModelFit est.
Llama 3.2 3B Instruct3B	Q4_K_M	3.6 GB	~61 tok/s ModelFit est.
Granite 4.1 3B Instruct3B	Q4_K_M	3.0 GB	~62 tok/s ModelFit est.
Gemma 4 E2B2.3B	Q8_0	5.7 GB	~72 tok/s ModelFit est.
Gemma 2 2B Instruct2B	Q4_K_M	2.8 GB	~69 tok/s ModelFit est.
Qwen3.5 2B Instruct2B	Q4_K_M	2.8 GB	~72 tok/s ModelFit est.
LFM3 1.7B1.73B	Q4_K_M	2.1 GB	~119 tok/s ModelFit est.
Qwen2.5 1.5B Instruct1.5B	Q4_K_M	2.5 GB	~123 tok/s ModelFit est.
Llama 3.2 1B Instruct1B	Q4_K_M	2.0 GB	~139 tok/s ModelFit est.
Gemma 3 1B Instruct1B	Q8_0	2.0 GB	~123 tok/s ModelFit est.
Qwen3.5 0.8B Instruct0.8B	Q4_K_M	1.8 GB	~149 tok/s ModelFit est.
Qwen2.5 0.5B Instruct0.5B	Q4_K_M	1.8 GB	~235 tok/s ModelFit est.
SmolLM2 360M0.36B	Q4_K_M	1.5 GB	~144 tok/s ModelFit est.

Speeds come from ModelFit's estimates (CC BY 4.0) where it covers the model, otherwise from our bandwidth formula. Method

Fits only at 3-bit or lower

Llama 3.3 Nemotron Super 49B.

Too big for this machine

DeepSeek-R1 Distill Qwen 32B (21 GB), Gemma 4 31B (21 GB), Qwen3.5 35B-A3B Instruct (21 GB), Qwen3 32B (22 GB), Laguna XS 2.1 (22 GB), Nanbeige4.3 33B (22 GB), Granite Switch 4.1 30B (23 GB), Gemma 2 27B Instruct (23 GB), Gemma 3 27B Instruct (23 GB), Ornith 1.0 35B (23 GB), Sage-A1.5 (23 GB), Qwen3.6-35B-A3B (24 GB) and more.

Questions

What is the best AI model for the M2 Pro (32 GB)?

Qwen3-Coder-30B-A3B at Q4_K_M: the largest model on our shortlist of current models that runs at 4-bit or better at a usable speed (about 45 tok/s, ModelFit est.).

How much memory can a model use on the M2 Pro (32 GB)?

About 21 GB of the 32 GB.

How fast is the M2 Pro (32 GB) for local LLMs?

Decode speed is limited by memory bandwidth: 200 GB/s. A model that reads 10 GB per token runs at under 20 tokens per second.