RTX 5080 × Llama 3.2 1B Instruct
Llama 3.2 1B Instruct on RTX 5080
103best gen tok/s
13,606best prompt tok/s
640max context tested
5results
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 | 103 | 13,606 | LocalScore ↗ | 512+128 | 96 ms | llamafile 0.9.2 |
| Q4_K - Medium | 41.7 | 9,611 | LocalScore ↗ | 512+128 | 144 ms | llamafile 0.9.3 |
| Q4_K - Medium | 41.2 | 9,533 | LocalScore ↗ | 512+128 | 143 ms | llamafile 0.9.3 |
| Q4_K - Medium | 41.1 | 9,524 | LocalScore ↗ | 512+128 | 144 ms | llamafile 0.9.3 |
| Q4_K - Medium | 41.0 | 9,253 | LocalScore ↗ | 512+128 | 145 ms | llamafile 0.9.3 |
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.
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.