Gemini 3 Pro

GoogleReleased 18 Nov 2025ClosedFlagshipDate verified

Compare Announcement

161150–172Index score#39 of 229 modelsDeep Think mode: 176

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

FrontierVersus the frontier at releaseset or matched the recordReplaced by Gemini 3.1 Pro

–Price per million tokensno public API price found

–Context windowtokens it reads at once

Where it sitson the frontier

Among the first models to reach this score.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence

GPQA Diamond

93%

Humanity's Last Exam

38%

Mathsevidence

FrontierMath T1-3

no result

Codingevidence

SWE-bench Verified

73%

Terminal-Bench

69%

Agentsevidence

APEX-Agents

no result

Novel problemsevidence

ARC-AGI-2

31%

Long tasksevidence

METR time horizon

3.7 h

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#39of 219 · 161

LMArena · Creative writingcreative writing#7of 386 · 1484 LMArena · Textoverall chat quality, judged by people#18of 388 · 1485 LMArena · Visionunderstanding images#11of 149 · 1289 LMArena · Mathsmaths questions in chat#36of 373 · 1477 LMArena · Codingcoding questions in chat#39of 383 · 1518 Epoch Capabilities Indexoverall capability from many benchmarks#38of 239 · 152.9 LMArena · WebDevbuilding web apps and front ends#63of 123 · 1439 LMArena · Documentsworking with long documents#31of 41 · 1434

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.