M2 (10-core GPU, 24 GB)
3 records across 3 models (quants: Q4_K - Medium).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M2 (10-core GPU, 24 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 14.668569888888888 | 141.73761555555558 | – | 9.023242666666668 | – | – | sourced | – | source |
| M2 (10-core GPU, 24 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 54.7565285 | 813.7094306666667 | – | 1.5797905915 | – | – | sourced | – | source |
| M2 (10-core GPU, 24 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 7.345236666666668 | 70.25649544444445 | – | 18.027677185222224 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085