Models /
Qwen3 4B Thinking 2507
3results
3devices
4.41Bparams
45.3fastest gen tok/s
llama.cpp
| Hardware | Quant | Gen tok/s | Prompt tok/s | Source | Context tested | Build | |
|---|---|---|---|---|---|---|---|
| RTX 4050 Laptop GPU | Q4_K - Medium | 45.3 | 1,806 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| RTX 4070 Laptop GPU | Q8_0 | 41.0 | 2,865 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| GTX 1060 6GB | Q4_K - Medium | 24.6 | 542 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
Ollama
No measured Ollama results yet. Submit yours →
MLX
No measured MLX results yet. Submit yours →
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.