RTX 3060 Laptop GPU × Llama 3.2 1B Instruct
Llama 3.2 1B Instruct on RTX 3060 Laptop GPU
210best gen tok/s
6,386best prompt tok/s
640max context tested
12results
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 | 210 | 6,386 | LocalScore ↗ | 512+128 | 223 ms | llamafile 0.9.2 |
| Q4_K - Medium | 181 | 5,836 | LocalScore ↗ | 512+128 | 249 ms | llamafile 0.9.2 |
| Q4_K - Medium | 162 | 6,037 | LocalScore ↗ | 512+128 | 227 ms | llamafile 0.9.3 |
| Q4_K - Medium | 160 | 6,037 | LocalScore ↗ | 512+128 | 231 ms | llamafile 0.9.3 |
| Q4_K - Medium | 160 | 5,174 | LocalScore ↗ | 512+128 | 281 ms | llamafile 0.9.2 |
| Q4_K - Medium | 157 | 5,787 | LocalScore ↗ | 512+128 | 236 ms | llamafile 0.9.3 |
| Q4_K - Medium | 153 | 6,016 | LocalScore ↗ | 512+128 | 231 ms | llamafile 0.9.3 |
| Q4_K - Medium | 148 | 5,276 | LocalScore ↗ | 512+128 | 267 ms | llamafile 0.9.3 |
| Q4_K - Medium | 136 | 4,902 | LocalScore ↗ | 512+128 | 276 ms | llamafile 0.9.2 |
| Q4_K - Medium | 135 | 4,908 | LocalScore ↗ | 512+128 | 280 ms | llamafile 0.9.2 |
| Q4_K - Medium | 126 | 5,096 | LocalScore ↗ | 512+128 | 268 ms | llamafile 0.9.2 |
| Q4_K - Medium | 86.4 | 3,700 | LocalScore ↗ | 512+128 | 322 ms | llamafile 0.9.2 |
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.
What else can RTX 3060 Laptop GPU run? →
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.