Models /
Llama 3.2 3B Instruct
14results
7devices
3.61Bparams
82.6fastest gen tok/s
llama.cpp
| Hardware | Quant | Gen tok/s | Prompt tok/s | Source | Context tested | Build | |
|---|---|---|---|---|---|---|---|
| AMD Radeon RX 9070 XT | Q3_K - Large | 82.6 | 1,046 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 9070 XT | Q3_K - Large | 79.4 | 1,019 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| RTX 4050 Laptop GPU | Q4_K - Medium | 60.0 | 2,518 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| AMD Radeon RX 7800 XT | Q4_K - Small | 57.1 | 1,203 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| AMD Radeon RX 6900 XT | Q4_K - Medium | 55.3 | 2,159 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q5_K - Medium | 53.6 | 2,339 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q6_K | 52.8 | 2,124 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| AMD Radeon RX 6900 XT | Q8_0 | 49.1 | 2,680 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| GTX 1060 6GB | Q4_K - Medium | 34.0 | 743 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| GTX 1060 6GB | Q4_K - Medium | 34.0 | 703 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| AMD Radeon RX 6900 XT | F16 | 29.1 | 1,475 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| GTX 1060 6GB | Q6_K | 27.9 | 682 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| GTX 1660 SUPER | Q4_K - Medium | 14.4 | 293 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| AMD Radeon 780M Graphics | Q4_K - Medium | 6.4 | 288 | 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.