Compare
Pick two or more models to compare tok/s across GPUs and overall composite scores. With no picks, shows every model seen.
| Model | A100 tok/s | A6000 tok/s | H100 tok/s | Composite |
|---|---|---|---|---|
| llama3.1:8b | 17.2 | 12.5 | 192.3 | 92.44 |
| llama3.2:1b | 47.5 | 44.3 | 570.5 | — |
| llama3.3:70b | — | 0.0 | — | — |
| qwen2.5:1.5b | 38.0 | 59.9 | 185.4 | 93.66 |
| qwen2.5:14b | 11.6 | 16.1 | 105.5 | 92.08 |
| qwen2.5:32b | — | 1.4 | — | — |
| qwen2.5:72b | — | — | 28.4 | — |
| qwen3:30b | — | 2.8 | — | — |