Llama 2 7B › NVIDIA GTX 1070 Ti
Llama 2 7B on the NVIDIA GTX 1070 Ti
Best measured: 43.4 tokens/s generation and 293 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 | 43.4 | 293 | tg128 | 860a9e4 eGPU | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | Vulkan FA off | 42.9 | 298 | tg128 | 860a9e4 eGPU | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | CUDA FA on | 37.9 | 791 | tg128 | 79c1160 @pebaryan | llama.cpp CUDA scoreboard (#15013) |
| Q4_0 | CUDA FA off | 37.8 | 714 | tg128 | 79c1160 @pebaryan | 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.
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 GTX 1070 Ti · Llama 2 7B on other hardware