GPUs & Macs › AMD Radeon Pro W6800X
AMD Radeon Pro W6800X: local LLM speed
Llama 2 7B (Q4_0) generates 86.5 tokens/s on the AMD Radeon Pro W6800X (Vulkan), and processes prompts at 511 tokens/s.
Measured results
| Model | Generate tok/s | Prompt tok/s | Quant | Backend | Test | Source |
|---|---|---|---|---|---|---|
| Llama 2 7B | 86.5 | 511 | Q4_0 | Vulkan FA off | tg128 | llama.cpp #10879 details |
What fits on the AMD Radeon Pro W6800X
Estimates for this chip are scaled from a measured run: Llama 2 7B Q4_0 at 86.5 tok/s (Vulkan, source). Fit assumes the whole model in GPU memory with ~8K context.
Memory size unknown for this device.
Similar hardware
AMD Radeon Pro W6800X Duo · AMD Radeon Pro V540 · AMD Radeon Pro W5700 · AMD FirePro S9300 x2 · AMD Radeon Pro V340 · AMD Radeon RX 470 · AMD Radeon Pro W5500 · AMD Radeon RX 480