M2 (16 GB) — benchmark record
M2 (16 GB) local LLM benchmarks on Apple Silicon Mac with 4 published rows across 4 models. Peak published speed is 58.0 tok/s on DeepSeek R1 Distill Llama 8B. Published runtimes include MLX, Ollama.
Quick take
Best published speed here is 58.0 tok/s on DeepSeek R1 Distill Llama 8B at Q4_K - Medium. This page is evidence, not the full buying answer.
Based on 4 external benchmarks; no lab runs yet.
Published runtimes: MLX, Ollama.
Current published coverage
Published model coverage includes DeepSeek R1 Distill Llama 8B, Qwen 3 8B, Gemma 3 12B, Phi-4 14B. Published runtime coverage on this chip includes MLX, Ollama. Fastest published row is 58.0 tok/s on DeepSeek R1 Distill Llama 8B at Q4_K - Medium.
This chip is part of a family. Compare all M2 RAM variants →
Raw benchmark rows for M2 (16 GB)
Rows sorted by avg tok/s descending. Click source badge to see original measurement page.
| Model | Quant | Avg tok/s | Runtime | Source |
|---|---|---|---|---|
| DeepSeek R1 Distill Llama 8B | Q4_K - Medium | 58.0 tok/s | Ollama | ref |
| Qwen 3 8B | Q4_K - Medium | 55.0 tok/s | Ollama | ref |
| Gemma 3 12B | Q4_K - Medium | 32.0 tok/s | Ollama | ref |
| Phi-4 14B | Q4_K - Medium | 28.0 tok/s | MLX | ref |
Models tested on this chip
Next step
Use Rankings when you need the best overall answer for a Mac, and keep this page open when you need the evidence behind that recommendation.
If you are comparing local Apple Silicon against rented cloud hardware, use AI Data Center Index for current GPU rental context.
Data
benchmarks.json — full dataset · chips.json — chip summaries · benchmarks.csv — CSV export
Data from in-house lab measurements plus community-published benchmarks. See all chip families →