M4 Max (40-core GPU, 128 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M4 Max (40-core GPU, 128 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 51.63987292592592 | 617.642594037037 | – | 1.9420189397222225 | – | – | sourced | – | source |
| M4 Max (40-core GPU, 128 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 182.5577164513889 | 3833.843851847222 | – | 0.30469161659027777 | – | – | sourced | – | source |
| M4 Max (40-core GPU, 128 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 28.74764654814815 | 326.6194374666667 | – | 3.7867302675777776 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085