RTX 3060 Laptop GPU × Meta Llama 3.1 8B Instruct
Meta Llama 3.1 8B Instruct on RTX 3060 Laptop GPU
45.5best gen tok/s
1,450best prompt tok/s
640max context tested
9results
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 | 45.5 | 1,439 | LocalScore ↗ | 512+128 | 890 ms | llamafile 0.9.3 |
| Q4_K - Medium | 45.2 | 1,450 | LocalScore ↗ | 512+128 | 899 ms | llamafile 0.9.3 |
| Q4_K - Medium | 45.1 | 1,410 | LocalScore ↗ | 512+128 | 913 ms | llamafile 0.9.3 |
| Q4_K - Medium | 44.5 | 1,426 | LocalScore ↗ | 512+128 | 944 ms | llamafile 0.9.3 |
| Q4_K - Medium | 43.5 | 1,265 | LocalScore ↗ | 512+128 | 1,065 ms | llamafile 0.9.2 |
| Q4_K - Medium | 39.8 | 1,250 | LocalScore ↗ | 512+128 | 1,056 ms | llamafile 0.9.3 |
| Q4_K - Medium | 38.4 | 1,094 | LocalScore ↗ | 512+128 | 1,227 ms | llamafile 0.9.2 |
| Q4_K - Medium | 34.8 | 1,050 | LocalScore ↗ | 512+128 | 1,153 ms | llamafile 0.9.2 |
| Q4_K - Medium | 30.0 | 891 | LocalScore ↗ | 512+128 | 8,221 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.