RTX 5070 + Ryzen 5 5600GT
5 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTX 5070 + Ryzen 5 5600GT | Qwen3.8-Flash-Next 177B | UD-IQ3_XXS | llama.cpp (CUDA) | 14 | – | – | – | – | 2026-10-05 | community | – | source |
| RTX 5070 + Ryzen 5 5600GT | Qwen3.8-Flash-Next 177B | UD-IQ3_XXS | llama.cpp (CUDA) | 11.5 | – | – | – | – | 2026-10-05 | community | – | source |
| RTX 5070 + Ryzen 5 5600GT | Qwen3.8-Flash-Next 177B | UD-IQ3_XXS | llama.cpp (CUDA) | 10.41 | – | – | – | tg10000 | 2026-10-05 | community | – | source |
| RTX 5070 + Ryzen 5 5600GT | Qwen3.8-Flash-Next 177B | UD-IQ3_XXS | llama.cpp (CUDA) | 10.15 | – | – | – | tg4892 | 2026-10-05 | community | – | source |
| RTX 5070 + Ryzen 5 5600GT | Qwen3.8-Flash-Next 177B | UD-IQ3_XXS | llama.cpp (CUDA) | 7 | – | – | – | – | 2026-10-05 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085