qwen2 14B
2 records across 2 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Radeon RX 9070 XT | qwen2 14B | Q4_0 | llama.cpp (Vulkan) | 58.78 | 2039.45 | – | – | tg128 | 2025-04-18 | community | – | source |
| A770 | qwen2 14B | Q4_0 | llama.cpp (Vulkan) | 31.09 | 647.35 | – | – | tg128 | 2025-04-18 | community | – | source |