Llama 2 7B on Tesla K80
14.3best gen tok/s
133best prompt tok/s
640max context tested
4results
Measured · all numbers below come from the linked sources
llama.cpp
| Quant | Gen tok/s | Prompt tok/s | Source | Context tested | TTFT | Build |
|---|---|---|---|---|---|---|
| Q4_0 | 14.3 | 133 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @32732f2 |
| Q4_0 | 13.8 | 133 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @32732f2 |
| Q4_0 | 9.5 | 88.3 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @5d46bab — Running on single GPU |
| Q4_0 | 9.4 | 89.5 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @5d46bab — Running on single GPU |
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.