supermarket

Vision-language

GPT-5.1

By OpenAI, in the GPT-5 family.

Released
12 Nov 2025
Status
Legacy
Context
400K
Max output
128K
Input per 1M tokens
$1.25
Output per 1M tokens
$10.00
Open weights
No
License
proprietary
Parameters
—
Knowledge cutoff
30 Sep 2024
Reasoning
Yes
Modalities
text, image → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Knowledge & reasoning41.3333 of 4
Coding & software engineering13.7Not ranked1 of 5
Agentic tool use42.8Not ranked1 of 2
Vision & multimodal understanding38.1Not ranked1 of 3
Factuality & hallucination48.0271 of 1
Human preference47.3354 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Chess PuzzlesPuzzles solved32.0%Independenthigh8 Dec 2025Epoch AI
Chess PuzzlesPuzzles solved17.0%Independentnone7 Aug 2026Epoch AI
Chess PuzzlesPuzzles solved14.0%Independentlow7 Aug 2026Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)90.1%Third partyhigh—https://conceptualreasoning.ai/dtbench via Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.4389Third partyhigh—https://conceptualreasoning.ai/lmca via Epoch AI
Mystery Game PuzzlesPuzzles solved19.0%Independentlow27 Aug 2026Epoch AI
Mystery Game PuzzlesPuzzles solved16.0%Independentmedium27 Aug 2026Epoch AI
Mystery Game PuzzlesPuzzles solved15.0%Independentnone27 Aug 2026Epoch AI
SimpleBenchScore (AVG@5)53.2%Third partyhigh—https://lmcouncil.ai/benchmarks via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy87.6%Independenthigh13 Nov 2025Epoch AI
GPQA DiamondAccuracy85.0%Independentmedium17 Nov 2025Epoch AI
GPQA DiamondAccuracy66.7%Independentnone7 Aug 2026Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
FrontierMath Tier 4 (v1)Accuracy12.5%Independenthigh13 Nov 2025Epoch AI
FrontierMath Tier 4 (v1)Accuracy4.2%Independentmedium17 Nov 2025Epoch AI
FrontierMath Tiers 1-3 (v1)Accuracy31.0%Independenthigh13 Nov 2025Epoch AI
FrontierMath Tiers 1-3 (v1)Accuracy26.9%Independentmedium17 Nov 2025Epoch AI
FrontierMath Tiers 1-3 (v1)Accuracy17.3%Independentlow25 Nov 2025Epoch AI
FrontierMath Tiers 1-3 (v1)Accuracy2.1%Independentnone25 Nov 2025Epoch AI
OTIS Mock AIME 2024-2025Accuracy88.6%Independenthigh13 Nov 2025Epoch AI
OTIS Mock AIME 2024-2025Accuracy85.6%Independentmedium17 Nov 2025Epoch AI
OTIS Mock AIME 2024-2025Accuracy63.9%Independentlow25 Nov 2025Epoch AI
OTIS Mock AIME 2024-2025Accuracy37.8%Independentnone7 Aug 2026Epoch AI

Coding & software engineering

Coding & software engineering results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GSOOPT@113.7%Third partyunknown—https://gso-bench.github.io/index.html via Epoch AI
GSOOPT@113.7%Third partyhigh—GSO Leaderboard via Epoch AI
SWE-bench Verified% resolved68.0%Independenthigh18 Feb 2026Epoch AI
SWE-bench Verified% resolved65.9%Independenthigh17 Feb 2026Epoch AI
Terminal-Bench 2.0Accuracy (mean over trials)47.6%Third partymedium—Terminal-Bench v2 Leaderboard via Epoch AI
Terminal-Bench 2.0Accuracy (mean over trials)47.6%Third partymedium—Terminal-Bench v2 Leaderboard via Epoch AI
Terminal-Bench 2.0Accuracy (mean over trials)47.6%Third partyunknown—https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI
WeirdML (v2)Average accuracy60.8%Third partyhigh—https://htihle.github.io/weirdml.html via Epoch AI

Agentic tool use

Agentic tool use results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Deep Research Bench (FutureSearch)Average score42.8%Third partylow—https://evals.futuresearch.ai/#drb via Epoch AI

Long context

Long context results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
CL-benchTasks solved23.7%Source not statedhigh—Epoch AI via Epoch AI
CL-benchTasks solved21.1%Source not statedunknown—Epoch AI via Epoch AI
CL-bench LifeTasks solved17.3%Source not statedhigh—Epoch AI via Epoch AI

Vision & multimodal understanding

Vision & multimodal understanding results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Visual Physics Comprehension Test (VPCT)Correct58.7%Third partyhigh—VPCT leaderboard via Epoch AI
Visual Physics Comprehension Test (VPCT)Correct53.3%Third partymedium—VPCT leaderboard via Epoch AI

Factuality & hallucination

Factuality & hallucination results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
SimpleQA VerifiedAccuracy (% correct)48.0%Independenthigh27 Aug 2026Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1394Independentmedium25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1341Independent—25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes10720Independent—25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes4915Independentmedium25 Sep 2026Arena
Document Arena (arena.ai)Arena score (Elo scale)1416Independent—13 Sep 2026Arena
Document Arena (arena.ai)Votes8244Independent—13 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1456Independenthigh25 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1438Independent—25 Sep 2026Arena
Text Arena (arena.ai)Votes44329Independent—25 Sep 2026Arena
Text Arena (arena.ai)Votes41596Independenthigh25 Sep 2026Arena
Vision Arena (arena.ai)Arena score (Elo scale)1251Independenthigh28 Sep 2026Arena
Vision Arena (arena.ai)Arena score (Elo scale)1240Independent—28 Sep 2026Arena
Vision Arena (arena.ai)Votes10798Independent—28 Sep 2026Arena
Vision Arena (arena.ai)Votes9843Independenthigh28 Sep 2026Arena

Other GPT-5 versions

Sources