2× RTX 3090 + Tesla V100
1 records across 1 models (quants: Q4_K_M).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2× RTX 3090 + Tesla V100 | Apertus v1.5 70B | Q4_K_M | llama.cpp | 17 | 81 | – | – | 65536 | 2026-08-03 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085