gpt-oss 20B MXFP4 MoE
2 records across 1 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTX 3060 | gpt-oss 20B MXFP4 MoE | as-published | llama.cpp (Vulkan) | 18.65 | 324.21 | – | – | tg128 | 2025-11-18 | community | – | source |
| RTX 3060 | gpt-oss 20B MXFP4 MoE | as-published | llama.cpp (Vulkan) | 16.18 | 184.9 | – | – | tg128 | 2025-11-18 | community | – | source |