Best local AI for Apple M2 Pro · 32GB

The best local AI on a Apple M2 Pro · 32GB right now is DeepSeek-R1 (distill) (8B (Llama) @ Q5_K_M, ~17–30 tok/s). Full ranking below — capability × real speed on this hardware, never sponsored.

#modelbest fitverdictspeed
1 DeepSeek-R1 (distill) 8B (Llama) Q5_K_M ✅ Runs comfortably 17–30 tok/s
2 Mistral Small 3 24B Q3_K_M ✅ Runs comfortably 9–15 tok/s
3 Qwen3 8B Q5_K_M ✅ Runs comfortably 16–25 tok/s
4 Llama 3.1 8B Q5_K_M ✅ Runs comfortably 17–30 tok/s
5 Qwen2.5 7B Q6_K ✅ Runs comfortably 17–30 tok/s
6 Gemma 4 12B Q6_K ✅ Runs comfortably 8–14 tok/s
7 Gemma 3 27B Q3_K_M ✅ Runs (tight) 6–10 tok/s
8 Mistral 7B 7B Q6_K ✅ Runs comfortably 16–25 tok/s
9 Gemma 2 27B Q3_K_M ✅ Runs (tight) 7–11 tok/s
10 Phi-4 14B Q5_K_M ✅ Runs comfortably 9–15 tok/s
11 Mistral Nemo 12B Q8_0 ✅ Runs comfortably 8–13 tok/s
12 Llama 3.2 3B Q4_K_M ✅ Runs comfortably 40–65 tok/s
$npx runlocal-sh install deepseek-r1

Rankings use the same GQA-aware fit + bandwidth-calibrated speed math as the CLI — no sponsorships, no affiliate re-sorting, ever.

Different machine? Apple M1 · 8GB · Apple M1 Pro · 16GB · Apple M2 · 16GB · Apple M3 Pro · 18GB · Apple M4 Pro · 48GB · Apple M3 Max · 36GB

Or check a specific model on your exact rig: open the checker →