Claude 4 Opus Achieves New SWE-bench Record with 73.20% Score

May 22, 2025

SWE-bench Leaderboard

update

Related Tool:

Claude Code

general

Anthropic's Claude 4 Opus has set a new benchmark record on SWE-bench Verified, achieving 73.20% and establishing itself as the leading autonomous coding model.

AI Power Rankings Impact

Ranking Impact:

•
Velocity+30%

Quantitative Data

73.20%

SWE-bench Performance

查看来源查看工具

Note: This analysis was compiled by AI Power Rankings based on publicly available information. Metrics and insights are extracted to provide quantitative context for tracking AI tool developments.

加载中...

Claude 4 Opus Achieves New SWE-bench Record with 73.20% Score - Claude Code

APR

[TRANSLATE] Navigation

[TRANSLATE] Quick Links

[TRANSLATE] Categories

[TRANSLATE] Features

Claude 4 Opus Achieves New SWE-bench Record with 73.20% Score

AI Power Rankings Impact

Ranking Impact:

Quantitative Data

Claude 4 Opus Achieves New SWE-bench Record with 73.20% Score

AI Power Rankings Impact

Ranking Impact:

Quantitative Data