Llama 2 7B on RTX A4000
85.2best gen tok/s
2,808best 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 | 85.2 | 2,808 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @65349f2 |
| Q4_0 | 83.8 | 2,684 | llama.cpp (CUDA) ↗ | 512+128 | — | llama.cpp CUDA @65349f2 |
| Q4_0 | 76.1 | 2,483 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @f5245b5 — coopmat2 |
| Q4_0 | 73.7 | 2,248 | llama.cpp (Vulkan) ↗ | 512+128 | — | llama.cpp Vulkan @f5245b5 — coopmat2 |
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.