Llama 3.3 8B
2 records across 1 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3x RTX PRO 4000 Blackwell | Llama 3.3 8B | unspecified | CUDA | – | 12615 | – | – | – | 2026-10 | community | – | source |
| 3x RTX PRO 4000 Blackwell | Llama 3.3 8B | unspecified | CUDA | – | 12583 | – | – | – | 2026-10 | community | – | source |