community pool / model answer

gpt-oss-20b-GGUF

chat · unsloth/gpt-oss-20b-GGUF

93.4 tok/s · model median · 3 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
M5 Pro 24GB4-bit69.5 tok/sreported415- GB4,224llama.cpp
RTX 3060 12GB ×24-bit93.4 tok/sreported-12.2 GB65,536llama.cpp
RTX 3090 24GB4-bit180.4 tok/sreported-- GB1,024llama.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.