TPSTokens Per Second

gpt-oss-20b

OpenAI🇺🇸 United States
active

Smaller open-weights sibling of gpt-oss-120b: 21B-param MoE (3.6B active), Apache 2.0, runs in 16GB of memory for on-device and edge use.

Compare with other models →
Context window131K tokens

Version History

gpt-oss-20b-launchmajor

gpt-oss-20b launches. Smaller open-weights sibling of gpt-oss-120b: 21B-param MoE (3.6B active), Apache 2.0, runs in 16GB of memory for on-device and edge use.

Benchmark Scores

Full leaderboard →
68.8%
GPQA
77.7%
LiveCodeBench
85.3%
MMLU
74.8%
MMLU-Pro
262.0 tokens_per_sec
Speed (tok/s)