gpt-oss-120b

OpenAIReleased 5 Aug 2025OpenDate verified

Compare Run it yourself →Announcement

131120–143Index score#114 of 229 models

4/ 8Core tests taken4 of 6 areas · plus 11 anchor tests

−15Versus the frontier at releasebest then: o3

$0.037/ $0.17Price per million tokensinput / output, list price

131KContext windowtokens it reads at once

Where it sitson the frontier

o1 reached this score 8 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence

GPQA Diamond

76%

Humanity's Last Exam

no result

Mathsevidence

FrontierMath T1-3

no result

Codingevidence

SWE-bench Verified

no result

Terminal-Bench

19%

Agentsevidence

APEX-Agents

Novel problemsno evidence yet

ARC-AGI-2

no result

Long tasksevidence

METR time horizon

42 min

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

What others sayrank on each leaderboard

The Model Indexindependent tests, our method#114of 219 · 131

LMArena · Mathsmaths questions in chat#162of 373 · 1381 Epoch Capabilities Indexoverall capability from many benchmarks#104of 239 · 139.9 Artificial Analysis Intelligence Indexoverall capability across ten evaluations#96of 207 · 12 LMArena · Textoverall chat quality, judged by people#190of 388 · 1352 LMArena · Codingcoding questions in chat#197of 383 · 1390 LMArena · Creative writingcreative writing#241of 386 · 1278

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.