GPT-5.6 Terra

OpenAIReleased 9 Jul 2026ClosedFlagship

Compare Announcement

178167–189Index score#10 of 229 models

4/ 8Core tests taken4 of 6 areas · plus 16 anchor tests

−7Versus the frontier at releasebest then: Claude Fable 5

$2.0/ $12Price per million tokensinput / output, list price

1.05MContext windowtokens it reads at once

Where it sitson the frontier

Claude Fable 5 reached this score 1 month earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence

GPQA Diamond

93%

Humanity's Last Exam

no result

Mathsevidence

FrontierMath T1-3

86%

Codingevidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agentsevidence

APEX-Agents

58%

Novel problemsevidence

ARC-AGI-2

84%

Long tasksno evidence yet

METR time horizon

no result

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#10of 219 · 178

Epoch Capabilities Indexoverall capability from many benchmarks#9of 239 · 159.8 Artificial Analysis Intelligence Indexoverall capability across ten evaluations#15of 207 · 42 LMArena · Mathsmaths questions in chat#33of 373 · 1479 LMArena · Codingcoding questions in chat#41of 383 · 1515 LMArena · Textoverall chat quality, judged by people#47of 388 · 1465 LMArena · Creative writingcreative writing#68of 386 · 1426 LMArena · Visionunderstanding images#31of 149 · 1268 LMArena · Documentsworking with long documents#12of 41 · 1472 LiveBench · Overalloverall, with fresh questions to limit contamination#18of 58 · 77.9 LMArena · WebDevbuilding web apps and front ends#43of 123 · 1519

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.