The Model Index
Up to four models

Claude Opus 4.1 vs Claude Opus 4

On the index (GPT-4 = 100), Claude Opus 4.1 scores 140 and Claude Opus 4 136. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Claude Opus 4.1
Presets
Claude Opus 4.1
140124–156
WeightsClosed
ReleasedAug 2025
Price in / out$15 / $75
Context200K
Core tests taken5 of 9
Claude Opus 4
136120–151
WeightsClosed
ReleasedMay 2025
Price in / out–
Context–
Core tests taken5 of 9

Claude Opus 4.1 and Claude Opus 4 are too close to call (4 points apart, within the uncertainty), ahead on 4 of 4 shared core tests.

Test by testthird-party results

Claude Opus 4.1Claude Opus 4
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)Humanity's Last ExamHumanity's Last ExamGPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMMMU ProMMMU ProSimpleBenchSimpleBenchVPCTVPCTMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH 500MATH 500MGSMMGSMFrontierMath (2025 set)FrontierMath (2025 set)FrontierMath tier 4 (2025 set)FrontierMath tier 4 (2025 set)CODINGSWE-bench VerifiedSWE-bench VerifiedLiveCodeBenchLiveCodeBenchWeirdMLWeirdMLAGENTSDTBenchDTBenchDeepResearch BenchDeepResearch BenchLMCALMCAPROFESSIONAL WORKLegalBenchLegalBenchMortgageTaxMortgageTaxTaxEval v2TaxEval v2
Every result as a table
TestClaude Opus 4.1Claude Opus 4
Reasoning & knowledge
GPQA Diamond (Epoch AI)77.3%76.3%
Humanity's Last Exam11.5%10.7%
GPQA Diamond (Vals AI)76.3%71.7%
MMLU Pro87.9%86.2%
MMMU Pro77.5%73.3%
SimpleBench60.0%58.8%
VPCT35.0%38.0%
Maths
AIME-style maths (OTIS mock)68.9%64.4%
MATH 50095.4%90.4%
MGSM94.4%93.8%
FrontierMath (2025 set)7.2%4.5%
FrontierMath tier 4 (2025 set)4.2%4.2%
Coding
SWE-bench Verified73.4%70.7%
LiveCodeBench66.5%70.2%
WeirdML45.9%43.7%
Agents
DTBench80.0%81.6%
DeepResearch Bench48.3%46.8%
LMCA37.1%37.4%
Professional work
LegalBench83.5%83.1%
MortgageTax61.1%58.6%
TaxEval v273.7%71.9%
Task length · METR40 sec38 sec

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG