Claude Opus 5 · SWE-bench Verified(Verified subset (500 problems)) 96% · 附件证据(ev-op5-card-swe-verified)
Claude Opus 5 的 SWE-bench Verified · Verified subset (500 problems) 官方出处。Adaptive thinking · max effort (card-wide default) · 96%。见《Claude Opus 5 System Card》System Card PDF p.153, §8.2, SWE-bench Verified paragraph。证据 ev-op5-card-swe-verified。
原文与定位
Claude Opus 5 · SWE-bench Verified(Verified subset (500 problems)),Adaptive thinking · max effort (card-wide default)配置,Pass rate 96%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.153, §8.2, SWE-bench Verified paragraph。 查看成绩条件
Claude Opus 5 achieved 96.0%. The paragraph states that each SWE-bench variant score is averaged over five trials and Verified is a 500-problem subset.
- 官方材料
- Claude Opus 5 System Card · PDF 第 153 页
- 定位
- 附件 · System Card PDF p.153, §8.2, SWE-bench Verified paragraph
- 材料版本
- Current saved PDF/TXT: 本地留存副本 and current-opus-5-system-card.txt; cover dated 2026-07-24; p.2 changelog records an update on 2026-08-19; later revision date, if any, unknown.
- 版本公布
- 未确认
- 披露时间
- 未确认 · 首发/后补未确认
- 采集时间
- 历史依据
- 暂无可靠历史版本依据
引用文本
Claude Opus 5 · SWE-bench Verified(Verified subset (500 problems)),Adaptive thinking · max effort (card-wide default)配置,Pass rate 96%(具体成绩公布时间未确认)。官方出处 Claude Opus 5 System Card(附件,Current saved PDF/TXT: 本地留存副本 and current-opus-5-system-card.txt; cover dated 2026-07-24; p.2 changelog records an update on 2026-08-19; later revision date, if any, unknown.)System Card PDF p.153, §8.2, SWE-bench Verified paragraph,https://www.anthropic.com/claude-opus-5-system-card#page=153。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-swe-verified/
关联成绩与条件
| 指标与成绩 | 配置与评测条件 | 证据关系 |
|---|---|---|
| Pass rate96%原文 96.0% · 单位 % 公布 未确认 | Adaptive thinking · max effort (card-wide default)
| 直接支持该成绩 在排名中查看此成绩 |