RTX 4070 Ti SUPER × Llama 3.2 1B Instruct
Llama 3.2 1B Instruct on RTX 4070 Ti SUPER
342best gen tok/s
17,080best prompt tok/s
640max context tested
10results
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 | 342 | 17,080 | LocalScore ↗ | 512+128 | 83 ms | llamafile 0.9.2 |
| Q4_K - Medium | 201 | 16,292 | LocalScore ↗ | 512+128 | 89 ms | llamafile 0.9.2 |
| Q4_K - Medium | 144 | 14,611 | LocalScore ↗ | 512+128 | 92 ms | llamafile 0.9.2 |
| Q4_K - Medium | 137 | 13,908 | LocalScore ↗ | 512+128 | 95 ms | llamafile 0.9.2 |
| Q4_K - Medium | 136 | 13,164 | LocalScore ↗ | 512+128 | 108 ms | llamafile 0.9.2 |
| Q4_K - Medium | 133 | 13,977 | LocalScore ↗ | 512+128 | 95 ms | llamafile 0.9.2 |
| Q4_K - Medium | 129 | 10,817 | LocalScore ↗ | 512+128 | 119 ms | llamafile 0.9.2 |
| Q4_K - Medium | 129 | 11,111 | LocalScore ↗ | 512+128 | 117 ms | llamafile 0.9.2 |
| Q4_K - Medium | 129 | 10,869 | LocalScore ↗ | 512+128 | 119 ms | llamafile 0.9.2 |
| Q4_K - Medium | 129 | 10,847 | LocalScore ↗ | 512+128 | 119 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 4070 Ti SUPER 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.