Llama 2 7B › NVIDIA Tesla P100
Llama 2 7B on the NVIDIA Tesla P100
Best measured: 66.5 tokens/s generation and 686 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 | 66.5 | 686 | tg128 | eec1e33 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | Vulkan FA off | 63.2 | 678 | tg128 | eec1e33 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | CUDA FA on | 62.0 | 787 | tg128 | b8372ee @Hedede | llama.cpp CUDA scoreboard (#15013) |
| Q4_0 | CUDA FA off | 58.4 | 761 | tg128 | b8372ee @Hedede | llama.cpp CUDA scoreboard (#15013) |
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.
Rent one by the hour
Rent a GPU on RunPodAffiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.
All results for the NVIDIA Tesla P100 · Llama 2 7B on other hardware