RTX 4060 Ti × Llama 3.2 1B Instruct
Llama 3.2 1B Instruct on RTX 4060 Ti
219best gen tok/s
9,463best 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 | 219 | 9,424 | LocalScore ↗ | 512+128 | 166 ms | llamafile 0.9.2 |
| Q4_K - Medium | 218 | 9,337 | LocalScore ↗ | 512+128 | 161 ms | llamafile 0.9.1 |
| Q4_K - Medium | 218 | 9,134 | LocalScore ↗ | 512+128 | 167 ms | llamafile 0.9.2 |
| Q4_K - Medium | 216 | 9,463 | LocalScore ↗ | 512+128 | 164 ms | llamafile 0.9.2 |
| Q4_K - Medium | 214 | 9,161 | LocalScore ↗ | 512+128 | 173 ms | llamafile 0.9.2 |
| Q4_K - Medium | 176 | 8,411 | LocalScore ↗ | 512+128 | 179 ms | llamafile 0.9.3 |
| Q4_K - Medium | 176 | 8,835 | LocalScore ↗ | 512+128 | 175 ms | llamafile 0.9.3 |
| Q4_K - Medium | 174 | 8,575 | LocalScore ↗ | 512+128 | 177 ms | llamafile 0.9.3 |
| Q4_K - Medium | 171 | 8,699 | LocalScore ↗ | 512+128 | 177 ms | llamafile 0.9.2 |
| Q4_K - Medium | 106 | 8,549 | LocalScore ↗ | 512+128 | 162 ms | llamafile 0.9.2 |
| Q4_K - Medium | 52.6 | 6,655 | LocalScore ↗ | 512+128 | 204 ms | llamafile 0.9.2 |
| Q4_K - Medium | 52.5 | 6,597 | LocalScore ↗ | 512+128 | 205 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 4060 Ti 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.