supermarket

Vision-language

Claude Opus 5

By Anthropic, in the Claude Opus family.

Released
24 Jul 2026
Status
Legacy
Context
1M
Max output
128K
Input per 1M tokens
$5.00
Output per 1M tokens
$25.00
Open weights
No
License
proprietary
Parameters
—
Knowledge cutoff
May 2026
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 & reasoning59.3104 of 4
Coding & software engineering63.5Not ranked2 of 5
Agentic tool use63.4Not ranked1 of 2
Computer & browser use31.411 of 1
Vision & multimodal understanding44.0Not ranked1 of 3
Factuality & hallucination59.9141 of 1
Human preference61.144 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 solved42.0%Independentmax24 Jul 2026Epoch AI
Chess PuzzlesPuzzles solved33.0%Independent—6 Aug 2026Epoch AI
Chess PuzzlesPuzzles solved20.0%Independentlow6 Aug 2026Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)97.9%Third partyxhigh—https://conceptualreasoning.ai/dtbench via Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)97.9%Third partyhigh—https://conceptualreasoning.ai/dtbench via Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)97.6%Third partymax—https://conceptualreasoning.ai/dtbench via Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)94.9%Third partymedium—https://conceptualreasoning.ai/dtbench via Epoch AI
DTBench (Decision Theory Benchmark)Accuracy (capability questions)93.1%Third partylow—https://conceptualreasoning.ai/dtbench via Epoch AI
Earthborne Rangers (EBR-bench)Objectives reached45.7%Independentmax5 Sep 2026Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.6445Third partyxhigh—https://conceptualreasoning.ai/lmca via Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.6367Third partyhigh—https://conceptualreasoning.ai/lmca via Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.633Third partymax—https://conceptualreasoning.ai/lmca via Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.6293Third partymedium—https://conceptualreasoning.ai/lmca via Epoch AI
LMCA (Language Model Conceptual Argumentation)Correlation with expert ratings (Pearson r)0.6258Third partylow—https://conceptualreasoning.ai/lmca via Epoch AI
Mystery Game PuzzlesPuzzles solved59.0%Independentmax25 Jul 2026Epoch AI
Mystery Game PuzzlesPuzzles solved37.0%Independent—6 Aug 2026Epoch AI
SimpleBenchScore (AVG@5)80.6%Third partyunknown—https://lmcouncil.ai/benchmarks via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy93.9%Independentmax24 Jul 2026Epoch AI
GPQA DiamondAccuracy92.9%Independent—6 Aug 2026Epoch AI
GPQA DiamondAccuracy87.9%Independentlow6 Aug 2026Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
FrontierMath Tier 4Accuracy73.2%Independentmax24 Jul 2026Epoch AI
FrontierMath Tiers 1-3Accuracy85.6%Independentmax24 Jul 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy98.9%Independentmax24 Jul 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy97.8%Independent—6 Aug 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy93.3%Independentlow6 Aug 2026Epoch AI

Coding & software engineering

Coding & software engineering results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
DeepSWEPass@173.6%Third partymax—https://deepswe.datacurve.ai/ via Epoch AI
DeepSWEPass@173.2%Third partyxhigh—https://deepswe.datacurve.ai/ via Epoch AI
DeepSWEPass@172.8%Third partyhigh—https://deepswe.datacurve.ai/ via Epoch AI
DeepSWEPass@168.9%Third partymedium—https://deepswe.datacurve.ai/ via Epoch AI
DeepSWEPass@158.1%Third partylow—https://deepswe.datacurve.ai/ via Epoch AI
FrontierCodeMain score (Mean@5)53.4%Third partymax—https://cognition.com/frontiercode via Epoch AI
PostTrainBenchAverage score35.0%Source not statedunknown—Epoch AI via Epoch AI
WeirdML (v2)Average accuracy91.8%Third partymax—https://htihle.github.io/weirdml.html via Epoch AI
WeirdML (v2)Average accuracy91.6%Third partyhigh—https://htihle.github.io/weirdml.html via Epoch AI
WeirdML (v2)Average accuracy86.3%Third party——https://htihle.github.io/weirdml.html via Epoch AI

Agentic tool use

Agentic tool use results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
BALROGAverage progress63.4%Third partymax—https://balrogai.com/ via Epoch AI

Computer & browser use

Computer & browser use results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
OSWorld 2.0Binary completion (500 steps)31.4%Third partymax—https://osworld-v2.xlang.ai/ via Epoch AI
OSWorld 2.0Binary completion (500 steps)30.2%Third partyxhigh—https://osworld-v2.xlang.ai/ via Epoch AI
OSWorld 2.0Binary completion (500 steps)29.0%Third partyhigh—https://osworld-v2.xlang.ai/ via Epoch AI
OSWorld 2.0Binary completion (500 steps)25.3%Third partymedium—https://osworld-v2.xlang.ai/ via Epoch AI
OSWorld 2.0Binary completion (500 steps)22.3%Third partylow—https://osworld-v2.xlang.ai/ via Epoch AI

Professional knowledge work

Professional knowledge work results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
APEX-AgentsPass@1 (all rubric criteria met)65.8%Source not statedmax—Epoch AI via Epoch AI

Vision & multimodal understanding

Vision & multimodal understanding results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Furniture AssemblyScore60.8%Independentmax10 Sep 2026Epoch AI

Factuality & hallucination

Factuality & hallucination results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
SimpleQA VerifiedAccuracy (% correct)59.9%Independentmax10 Aug 2026Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1693Independentmax25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Arena score (Elo scale)1662Independenthigh25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes18930Independenthigh25 Sep 2026Arena
Code Arena: WebDev (arena.ai)Votes15627Independentmax25 Sep 2026Arena
Document Arena (arena.ai)Arena score (Elo scale)1490Independenthigh13 Sep 2026Arena
Document Arena (arena.ai)Votes8794Independenthigh13 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1491Independenthigh25 Sep 2026Arena
Text Arena (arena.ai)Arena score (Elo scale)1488Independentmax25 Sep 2026Arena
Text Arena (arena.ai)Votes55063Independenthigh25 Sep 2026Arena
Text Arena (arena.ai)Votes26760Independentmax25 Sep 2026Arena
Vision Arena (arena.ai)Arena score (Elo scale)1289Independenthigh28 Sep 2026Arena
Vision Arena (arena.ai)Votes15304Independenthigh28 Sep 2026Arena

Other Claude Opus versions

Sources