Llama 2 7B on the NVIDIA RTX 5080
Best measured: 192 tokens/s generation and 8,055 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 | 192 | 8,055 | tg128 | f6b533d coopmat2 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | Vulkan FA off | 185 | 7,445 | tg128 | f6b533d coopmat2 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | CUDA FA on | 185 | 9,488 | tg128 | 8a4280c @Hedede | llama.cpp CUDA scoreboard (#15013) |
| Q4_0 | CUDA FA off | 182 | 8,297 | tg128 | 8a4280c @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.
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 RTX 5080 · Llama 2 7B on other hardware