RTX 4070 Ti × Llama 3.2 1B Instruct
Llama 3.2 1B Instruct on RTX 4070 Ti
309best gen tok/s
15,156best prompt tok/s
640max context tested
6results
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 | 309 | 15,156 | LocalScore ↗ | 512+128 | 99 ms | llamafile 0.9.2 |
| Q4_K - Medium | 200 | 15,093 | LocalScore ↗ | 512+128 | 101 ms | llamafile 0.9.2 |
| Q4_K - Medium | 199 | 14,206 | LocalScore ↗ | 512+128 | 104 ms | llamafile 0.9.2 |
| Q4_K - Medium | 189 | 14,676 | LocalScore ↗ | 512+128 | 103 ms | llamafile 0.9.2 |
| Q4_K - Medium | 55.0 | 8,761 | LocalScore ↗ | 512+128 | 154 ms | llamafile 0.9.2 |
| Q4_K - Medium | 41.4 | 9,013 | LocalScore ↗ | 512+128 | 173 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 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.