Llama 2 7B on GTX 1080 Ti
65.0best gen tok/s
1,138best 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 | 65.0 | 541 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @360d653 |
| Q4_0 | 64.6 | 530 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @360d653 |
| Q4_0 | 62.5 | 1,084 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @9c35706 |
| Q4_0 | 61.4 | 1,138 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @9c35706 |
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.