ModelBeat
Models/Claude Opus 4.7
All models
A

Anthropic: Claude Opus 4.7

anthropic/claude-opus-4.7

Claude Opus 4.7 is Anthropic’s proprietary AI model, released Apr 16, 2026. It costs $5/M input and $25/M output tokens, and handles a 1M-token context window. Its strongest use case is coding, where it ranks in the 94th percentile of the models Model Beat tracks.

Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.

Context
1M
Input
$5 / 1M
Output
$25 / 1M
Released
Apr 16, 2026
Modalities
textimagefiletext
Strongest atCoding top 6%Math top 14%Reasoning & Knowledge top 25%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Claude Opus 4.7 lands among all models tracked on Model Beat.

82.0
Intelligence Index
Epoch AI
82nd percentile of tracked models
94.0
Coding Index
Epoch AI
94th percentile of tracked models
73.0
Agentic Index
Epoch AI
73rd percentile of tracked models

Reasoning

7 evals
ARC-AGI93.5%

Abstract visual reasoning

ARC-AGI-275.8%

Harder abstract reasoning

GPQA Diamond90.2%

Graduate-level scientific reasoning

Humanity's Last Exam36.2%

Frontier of human expert knowledge

SimpleBench62.9%

Common-sense trick questions

SimpleQA Verified50.6%

Factual accuracy & hallucination

WeirdML76.4%

Novel ML problem-solving

Coding

4 evals
GSO (code optimization)44.1%

Code performance optimization

SciCode54.5%

Scientific research coding

SWE-bench Verified83.5%

Real GitHub issue resolution

WebDev Arena1567

Human-rated web development

Math

3 evals
AIME 2024/202597.8%

Olympiad-qualifier math

FrontierMath43.8%

Research-level math problems

FrontierMath Tier 422.9%

Hardest research math

Agentic & Tools

3 evals
APEX33.9%

Multi-step agentic tasks

Terminal-Bench80.2%

Command-line agentic tasks

τ²-bench88.6%

Tool-agent-user reliability

In the news

4

Changelog

1
  • Claude Opus 4.7: Terminal-Bench dropped from 90.2% to 80.2% (-11%)benchmark · Aug 5, 2026

Compare Claude Opus 4.7 with

Frequently asked questions

What is Claude Opus 4.7?

Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.

Who created Claude Opus 4.7?

Claude Opus 4.7 is an AI model developed by Anthropic, part of the Claude family, released Apr 16, 2026.

How much does Claude Opus 4.7 cost?

Claude Opus 4.7's list price from Anthropic is $5 per million input tokens and $25 per million output tokens.

What is Claude Opus 4.7's context window?

Claude Opus 4.7 supports a context window of up to 1M tokens.

How does Claude Opus 4.7 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Claude Opus 4.7 scores 83.5% on SWE-bench Verified, 44.1% on GSO (code optimization), 54.5% on SciCode.

Is Claude Opus 4.7 open source?

Claude Opus 4.7 is a proprietary model (API access), available through its provider's API rather than as a downloadable open-weight model.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Aug 5, 2026.