Models /
Mistral Nemo Instruct 2407
5results
2devices
12.25Bparams
63.1fastest gen tok/s
llama.cpp
| Hardware | Quant | Gen tok/s | Prompt tok/s | Source | Context tested | Build | |
|---|---|---|---|---|---|---|---|
| RTX A5000 | Q4_K - Medium | 63.1 | 1,959 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q4_K - Medium | 27.1 | 746 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q5_K - Medium | 25.4 | 717 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q6_K | 23.8 | 647 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q8_0 | 20.7 | 832 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | 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.