Apple M3 Pro (14-core GPU) × Llama 3.2 1B Instruct
Llama 3.2 1B Instruct on Apple M3 Pro (14-core GPU)
88.7best gen tok/s
1,344best prompt tok/s
640max context tested
4results
Measured · all numbers below come from the linked sources
llama.cpp
| Quant | Gen tok/s | Prompt tok/s | Source | Context tested | TTFT | Build |
|---|---|---|---|---|---|---|
| Q4_K - Medium | 88.7 | 1,243 | LocalScore ↗ | 512+128 | 993 ms | llamafile 0.9.2 |
| Q4_K - Medium | 88.3 | 1,333 | LocalScore ↗ | 512+128 | 955 ms | llamafile 0.9.2 |
| Q4_K - Medium | 88.2 | 1,323 | LocalScore ↗ | 512+128 | 952 ms | llamafile 0.9.2 |
| Q4_K - Medium | 88.1 | 1,344 | LocalScore ↗ | 512+128 | 959 ms | llamafile 0.9.2 |
Ollama
No measured Ollama results yet. Submit yours →
MLX
No measured MLX results yet. Submit yours →
"Context tested" is the largest prompt+generation length actually benchmarked (tokens), not a memory-fit estimate.
What else can Apple M3 Pro (14-core GPU) run? →
Includes results from LocalScore, a Mozilla Builders project by cjpais, mirrored in the cn0303/local-llm-speed-benchmarks dataset. LocalScore runs use llamafile, which is built on llama.cpp.