AIEvals

Models

Claude Opus 4.7

Specifications and API prices for Claude Opus 4.7, recorded by Artificial Analysis.

AnthropicReleased 16 April 2026Context 1M tokensOutput Not published in the checked sourcePrivate weights$5.00 in / $25.00 out per Mtok

Family Claude Opus33 benchmarks with a published result1 independent first place among the 46 models this site listsRead 17 October 2026

Where it stands

Rank compares tracked models with a score. Published shows the source’s rank and field size, when available.

BenchmarkScoreRankPublishedLeaderBehind
Artificial Analysis Intelligence IndexArtificial Analysis4123of 42no list readClaude Opus 5.517 pts
SWE-bench (Verified)Vals AI82.00%16of 2321of 88Claude Opus 515.00 pts
LiveCodeBenchVals AI85.07%16of 2136of 143Claude Fable 5.15.45 pts
GPQA DiamondVals AI90.15%18of 2325of 138Gemini 3.1 Pro5.30 pts
MMLU ProVals AI89.87%7of 248of 138Claude Fable 5.12.51 pts
MMMU ProVals AI85.55%15of 2024of 93Claude Fable 5.15.09 pts
Vibe Code Bench v1.1Vals AI71.00%28of 3832of 106Claude Sonnet 5.521.39 pts
ProgramBenchVals AI0.00%19of 2923of 54Claude Opus 5.518.50 pts
τ³-BankingArtificial Analysis34.64%23of 30no list readQwen3.8 Max16.70 pts
ITBench-AAArtificial Analysis46.66%10of 18no list readGPT-5.6 Sol9.55 pts
IFBenchArtificial Analysis58.64%10of 10no list readGrok 4 20 030924.29 pts
AA-AnalystAgentArtificial Analysis43.75%11of 18no list readGemini 3.7 Flash16.25 pts
MLCR-AAArtificial Analysis28.89%16of 37no list readClaude Sonnet 5.546.11 pts
Vals RSI IndexVals AI19.08%17of 2017of 23Claude Opus 5.518.23 pts
τ²-Bench TelecomArtificial Analysis88.60%7of 10no list readJt 35b Flash10.52 pts
CritPtArtificial Analysis12.00%37of 42no list readGPT-5.6 Sol20.29 pts
AA-OmniscienceArtificial Analysis27.2714of 42no list readClaude Opus 5.519.15
AA-LCR v1.1Artificial Analysis78.67%36of 42no list readKimi K310.00 pts
Terminal-Bench 2.1Vals AI68.54%26of 3531of 75Claude Opus 5.519.10 pts
LegalBenchVals AI85.25%15of 2722of 149Claude Fable 53.31 pts
Harvey's Legal Agent BenchmarkVals AI6.67%18of 3724of 73Muse Spark 1.218.75 pts
Vals Multimodal IndexVals AI67.36%10of 1710of 33Claude Fable 56.79 pts
Code MigrationVals AI43.88%22of 3722of 71Claude Sonnet 5.525.95 pts
EMBVals AI63.73%23of 3724of 69Claude Fable 5.112.94 pts
Finance Agent (v2)Vals AI51.51%28of 3734of 73Gemini 4 Argon13.89 pts
Legal Research BenchVals AI38.46%26of 3728of 72Muse Spark 1.3 Max16.83 pts
MedCodeVals AI54.86%5of 357of 104Claude Opus 58.71 pts
MedScribeVals AI82.95%30of 3647of 106Claude Opus 5.58.48 pts
SAGEVals AI56.10%1of 321of 90Claude Opus 4.7—
Tax Agent BenchVals AI63.86%19of 3722of 64Claude Fable 5.113.78 pts
GDPval-AA v2.1Artificial Analysis1,338.4236of 42no list readClaude Opus 5.5507.75
AA-Briefcase v1.1Artificial Analysis1,25334of 41no list readClaude Opus 5.5568.85
Humanity's Last ExamArtificial Analysis42.31%30of 42no list readClaude Opus 5.519.04 pts

Nearby on Artificial Analysis Intelligence Index

The two models above and below on the Artificial Analysis Intelligence Index (Artificial Analysis), on every column where this model has a published number.

ModelArtificial Analysis Intelligence IndexindexSWE-bench%LiveCodeBench%GPQA Diamond%MMLU Pro%MMMU Pro%Vibe Code Bench v1.1%ProgramBench%
GLM 5.3 FlashZ.ai4292.00%80.51%86.36%86.06%86.01%30.76%0.00%
GPT-5.6 TerraOpenAI4295.40%85.93%90.91%86.66%86.47%74.59%0.50%
Claude Opus 4.7Anthropic4182.00%85.07%90.15%89.87%85.55%71.00%0.00%
Gemini 3.8 FlashGoogle4180.00%89.48%94.44%90.22%89.08%78.65%1.00%
Muse Spark 1.2Meta4086.60%––88.28%86.13%79.10%–
Bold marks a figure better than this model’s on the same column, whichever model it belongs to. Ranks cover the models shown from the stated source.

Cost and speed, as published

Additional measurements from Artificial Analysis’ own row. Read 30 September 2026.

As publishedCost per taskOutput speed
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)—45.9 tok/s

Results by benchmark

Independent and vendor results retain their own harnesses and effort settings.

BenchmarkIndependent runAs reported by a publisher
τ³-BankingTool use34.64%Artificial Analysisnot published
ITBench-AAIT operations46.66%Artificial Analysisnot published
IFBenchInstruction following58.64%Artificial Analysisnot published
AA-AnalystAgentData analysis43.75%Artificial Analysisnot published
MLCR-AAMedical reasoning28.89%Artificial Analysisnot published
Vals RSI IndexAutonomous research19.08%Vals AInot published
τ²-Bench TelecomTool use88.60%Artificial Analysisnot published
CritPtScientific reasoning12.00%Artificial Analysisnot published
AA-OmniscienceFactual knowledge27.27Artificial Analysisnot published
AA-LCR v1.1Long-context reasoning78.67%Artificial Analysisnot published
Terminal-Bench 2.1Agentic coding68.54%Vals AI
83.15%Artificial Analysis
not published
SWE-bench (Verified)Software engineering82.00%Vals AInot published
LiveCodeBenchCode generation85.07%Vals AInot published
GPQA DiamondScience reasoning90.15%Vals AI
91.41%Artificial Analysis
not published
MMLU ProKnowledge89.87%Vals AInot published
MMMU ProMultimodal reasoning85.55%Vals AI
78.84%Artificial Analysis
not published
Vibe Code Bench v1.1Product coding71.00%Vals AInot published
ProgramBenchAgentic coding0.00%Vals AInot published
LegalBenchProfessional knowledge85.25%Vals AInot published
Harvey's Legal Agent BenchmarkProfessional knowledge6.67%Vals AInot published
Vals Multimodal IndexComposite67.36%Vals AInot published
Code MigrationAgentic coding43.88%Vals AInot published
EMBFinancial modelling63.73%Vals AInot published
Finance Agent (v2)Financial analysis51.51%Vals AInot published
Legal Research BenchProfessional knowledge38.46%Vals AInot published
MedCodeHealthcare administration54.86%Vals AInot published
MedScribeHealthcare administration82.95%Vals AInot published
SAGEEducation56.10%Vals AInot published
Tax Agent BenchProfessional knowledge63.86%Vals AInot published
GDPval-AA v2.1Knowledge work1,338.42Artificial Analysisnot published
AA-Briefcase v1.1Knowledge work1,253Artificial Analysisnot published
Humanity's Last ExamMultidisciplinary reasoning42.31%Artificial Analysisnot published
Artificial Analysis Intelligence IndexComposite41Artificial Analysisnot published
Independent column: the suite named under each figure, one harness per suite. Publisher column: each lab’s own table, reproduced on the benchmark’s page. Publisher ranks compare reported scores in the same table row only; ties share a rank and unreported cells are excluded.

Sources

Read on 17 October 2026.