Event ·Official first-hand

Claude 3.5 Sonnet 在 SWE-bench Verified 上刷新纪录

Anthropic 宣布其 Claude 3.5 Sonnet 模型在 SWE-bench Verified 基准测试中取得了 49.7% 的通过率，较此前最佳成绩提升了约 10 个百分点。该测试评估 AI 模型解决真实 GitHub 问题的能力，包括代码修复、功能实现等。Claude 3.5 Sonnet 在多个类别中表现优异，尤其在需要多步推理和上下文理解的复杂任务上。这一进展表明 AI 在软件工程自动化领域正快速接近人类水平。

Current read

Claude 3.5 Sonnet 在 SWE-bench 上的突破意味着 AI 编程助手离真正解决复杂工程问题更近了一步，做软件开发的团队可以关注这一能力提升对日常代码修复和功能开发的潜在影响。

11 sources75° AI heatLast updated 2026/5/12 22:33:23

Evidence

11 sources

Primary sourceAnthropic: Engineering

2026/5/12 22:33:23View source

Related source 1Claude: Blog

2026/5/13 04:33:11View source

Related source 2arXiv: OpenAI

2026/5/13 11:12:27View source

Related source 3宝玉

2026/5/13 19:55:08View source

Related source 4elvis

2026/5/13 21:46:47View source

Related source 5向阳乔木

2026/5/14 02:56:30View source

Related source 6AI Will

2026/5/14 07:59:13View source

Related source 7Ethan Mollick

2026/5/11 03:18:01View source

Related source 8Dario Amodei Blog

2026/5/12 17:58:43View source

Related source 9The Rundown AI

2026/5/13 01:11:45View source

Related source 10IT之家

2026/5/13 07:05:18View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story