RTX 4060 × Meta Llama 3.1 8B Instruct
Meta Llama 3.1 8B Instruct on RTX 4060
45.8best gen tok/s
1,748best prompt tok/s
4,352max 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 | 45.8 | 1,748 | LocalScore ↗ | 512+128 | 772 ms | llamafile 0.9.2 |
| Q4_K - Medium | 41.2 | 1,708 | LocalScore ↗ | 512+128 | 793 ms | llamafile 0.9.2 |
| Q4_K - Medium | 40.7 | 1,702 | LocalScore ↗ | 512+128 | 795 ms | llamafile 0.9.2 |
| Q4_K - Medium | 39.6 | 1,624 | LocalScore ↗ | 4,352 | 825 ms | llamafile 0.9.3 |
| Q4_K - Medium | 23.4 | 857 | LocalScore ↗ | 512+128 | 2,198 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.