community pool / model answer

gpt-oss-120b

chat · openai/gpt-oss-120b

70.7 tok/s · model median · 10 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Arc Pro B70 32GB ×24-bit59.4 tok/sreported-60 GB2,048llama.cpp
GB10 Grace Blackwell 128GB4-bit34.4 tok/sreported-- GB32,768vllm
Radeon AI Pro R9700 32GB ×316-bit69.9 tok/smeasured69168.5 GB845llama.cpp
RTX PRO 6000 Blackwell 96GB16-bit196 tok/smeasured585- GB15,964llama.cpp
Ryzen AI Max 395 128GB16-bit38.8 tok/sreported1,718- GB845llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.