GTX 1080 Ti
8 records across 1 models (quants: UD-Q8_K_XL).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 19.32 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 18.72 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 18.71 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 18.64 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 17.26 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 17.19 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 14.38 | – | – | – | 8192 | 2026-10-05 | community | – | source |
| GTX 1080 Ti | Qwen3.6-35B-A3B 35B-A3B | UD-Q8_K_XL | llama.cpp (CUDA) | 13.25 | – | – | – | 8192 | 2026-10-05 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085