gpt-oss-20b
gpt-oss-20b is an open-weights AI model from OpenAI with an index score of 128 (GPT-4 = 100), rank 137 of 217.
128112–144Index score#137 of 217 models
$0.018/ $0.090Price per million tokensinput / output · via OpenRouter
131KContext windowtokens it reads at once
5 Aug 2025Released1 year 3 months ago
Where it sitson the frontier
o1 reached this score 8 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 8 leaderboards
The Model Indexthird-party tests, our method#137of 217 · 128
LMArena · Mathsmaths questions in chat#183of 346 · 1335Epoch Capabilities Indexoverall capability from many benchmarks#132of 231 · 134.9LMArena · Codingcoding questions in chat#205of 357 · 1368LMArena · Textoverall chat quality, judged by people#222of 362 · 1318LMArena · Instruction followingfollowing detailed instructions#237of 362 · 1281LMArena · Long promptslong, complex requests#235of 340 · 1303LMArena · Creative writingcreative writing#253of 360 · 1241Best attop three among current models
Not in the top three among current models on any category we track.
Where to use itprices checked 18 Nov 2026
API providers 8 of 10 · per 1M tokens, in / out
Run it yourself
Official weightsopenai/gpt-oss-20b↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released5 Aug 2025
Announcementopenai.com ↗
Release entrygpt-oss-120b/20b
Scored asgpt-oss-20b
Output speed–
Hugging Faceopenai/gpt-oss-20b ↗