Llama 2 7B › NVIDIA Quadro P2000
Llama 2 7B on the NVIDIA Quadro P2000
Best measured: 23.8 tokens/s generation and 182 tokens/s prompt processing (Q4_0, Vulkan, FA on).
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| Q4_0 | Vulkan FA on | 23.8 | 182 | tg128 | 63f8fe0 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | CUDA FA on | 23.8 | 312 | tg128 | baa9255 @TinyServal | llama.cpp CUDA scoreboard (#15013) |
| Q4_0 | CUDA FA off | 23.6 | 309 | tg128 | baa9255 @TinyServal | llama.cpp CUDA scoreboard (#15013) |
| Q4_0 | Vulkan FA off | 23.1 | 170 | tg128 | 63f8fe0 | llama.cpp Vulkan scoreboard (#10879) |
tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128.
No card yet?
Try it on a rented GPU (RunPod)Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.
All results for the NVIDIA Quadro P2000 · Llama 2 7B on other hardware