Llama 2 7B › AMD Radeon RX 7900 GRE
Llama 2 7B on the AMD Radeon RX 7900 GRE
Best measured: 116 tokens/s generation and 2,336 tokens/s prompt processing (Q4_0, Vulkan, FA off).
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| Q4_0 | Vulkan FA off | 116 | 2,336 | tg128 | 4b2a477 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | Vulkan FA on | 110 | 2,251 | tg128 | 4b2a477 | llama.cpp Vulkan scoreboard (#10879) |
| Q4_0 | ROCm FA on | 97.5 | 1,599 | tg128 | 6fa3b55 @MihaiBojescu | llama.cpp ROCm scoreboard (#15021) |
| Q4_0 | ROCm FA off | 96.1 | 1,457 | tg128 | 6fa3b55 @MihaiBojescu | llama.cpp ROCm scoreboard (#15021) |
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 AMD Radeon RX 7900 GRE · Llama 2 7B on other hardware