RTX 5060 Ti × Meta Llama 3.1 8B Instruct
Meta Llama 3.1 8B Instruct on RTX 5060 Ti
62.6best gen tok/s
2,535best prompt tok/s
4,352max context tested
7results
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 | 62.6 | 2,387 | LocalScore ↗ | 512+128 | 539 ms | llamafile 0.9.2 |
| Q4_K - Medium | 61.9 | 2,495 | LocalScore ↗ | 512+128 | 537 ms | llamafile 0.9.2 |
| Q4_K - Medium | 61.5 | 2,535 | LocalScore ↗ | 512+128 | 526 ms | llamafile 0.9.2 |
| Q4_K - Medium | 60.6 | 2,427 | LocalScore ↗ | 512+128 | 557 ms | llamafile 0.9.2 |
| Q4_K - Medium | 55.3 | 2,235 | LocalScore ↗ | 4,352 | 2,767 ms | llamafile 0.9.2 |
| Q4_K - Medium | 53.1 | 2,089 | LocalScore ↗ | 512+128 | 664 ms | llamafile 0.9.2 |
| Q4_K - Medium | 41.1 | 2,163 | LocalScore ↗ | 512+128 | 599 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 5060 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.