Hardware /
RTX 2060
10measured results
168best gen tok/s*
6 GBreported memory
*Headline excludes tiny/test models (under 0.6B params or TinyStories-style). Open in rig checker →
llama.cpp
| Model | Quant | Gen tok/s | Prompt tok/s | Source | Context tested | Build | |
|---|---|---|---|---|---|---|---|
| Llama 3.2 1B Instruct | Q4_K - Medium | 168 | 4,875 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| Llama 3.2 1B Instruct | Q4_K - Medium | 128 | 4,640 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| Llama 3.2 1B Instruct | Q4_K - Medium | 125 | 4,618 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| Llama 3.2 1B Instruct | Q4_K - Medium | 125 | 3,784 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| Qwen2.5 Coder 7B Instruct | Q4_K - Medium | 41.0 | 1,159 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| Meta Llama 3.1 8B Instruct | Q4_K - Medium | 38.1 | 960 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| Gemma 3 4b It | Q6_K | 24.9 | 716 | LocalScore ↗ | 512+128 | llamafile 0.9.2 | Details → |
| Qwen2.5 14B Instruct | Q4_K - Medium | 23.4 | 600 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| Qwen2.5 14B Instruct | Q4_K - Medium | 22.2 | 589 | LocalScore ↗ | 512+128 | llamafile 0.9.3 | Details → |
| Mistral Small 22B ArliAI RPMax v1.1 | Q5_K - Medium | 2.0 | 110 | 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.