Threadripper PRO 3975WX + 2x RTX 5090
4 records across 1 models (quants: UD-IQ3_XXS).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Threadripper PRO 3975WX + 2x RTX 5090 | GLM-5.3-Flash | UD-IQ3_XXS | llama.cpp (CUDA) | 33.5 | – | – | – | 204800 | 2026-10-07 | community | – | source |
| Threadripper PRO 3975WX + 2x RTX 5090 | GLM-5.3-Flash | UD-IQ3_XXS | llama.cpp (CUDA) | 23.7 | – | – | – | 204800 | 2026-10-07 | community | – | source |
| Threadripper PRO 3975WX + 2x RTX 5090 | GLM-5.3-Flash | UD-IQ3_XXS | llama.cpp (CUDA) | 18.5 | – | – | – | 204800 | 2026-10-07 | community | – | source |
| Threadripper PRO 3975WX + 2x RTX 5090 | GLM-5.3-Flash | UD-IQ3_XXS | llama.cpp (CUDA) | 12.9 | – | – | – | 204800 | 2026-10-07 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085