Qwen3 4B on RTX 3090
117best gen tok/s
4,467best prompt tok/s
640max context tested
3results
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 | 117 | 4,467 | LocalScore ↗ | 512+128 | 298 ms | llamafile 0.9.3 |
| Q4_K - Medium | 66.1 | 3,933 | LocalScore ↗ | 512+128 | 341 ms | llamafile 0.9.3 |
| Q4_K - Medium | 65.2 | 4,397 | LocalScore ↗ | 512+128 | 302 ms | llamafile 0.9.3 |
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.
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.