GTX 1080 Ti × Qwen2.5 14B Instruct
Qwen2.5 14B Instruct on GTX 1080 Ti
21.1best gen tok/s
449best prompt tok/s
640max context tested
5results
Measured · all numbers below come from the linked sources
llama.cpp
| Quant | Gen tok/s | Prompt tok/s | Source | Context tested | TTFT | Build |
|---|---|---|---|---|---|---|
| Q4_K - Medium | 21.1 | 449 | LocalScore ↗ | 512+128 | 2,817 ms | llamafile 0.9.2 |
| Q4_K - Medium | 20.6 | 427 | LocalScore ↗ | 512+128 | 3,008 ms | llamafile 0.9.2 |
| Q4_K - Medium | 20.2 | 424 | LocalScore ↗ | 512+128 | 2,960 ms | llamafile 0.9.2 |
| Q4_K - Medium | 19.6 | 412 | LocalScore ↗ | 512+128 | 3,054 ms | llamafile 0.9.2 |
| Q4_K - Medium | 19.4 | 381 | LocalScore ↗ | 512+128 | 3,388 ms | llamafile 0.9.2 |
Ollama
No measured Ollama results yet. Submit yours →
MLX
No measured MLX results yet. Submit yours →
"Context tested" is the largest prompt+generation length actually benchmarked (tokens), not a memory-fit estimate.
What else can GTX 1080 Ti run? →
Includes results from LocalScore, a Mozilla Builders project by cjpais, mirrored in the cn0303/local-llm-speed-benchmarks dataset. LocalScore runs use llamafile, which is built on llama.cpp.