Claude Opus 4.6 · SWE-bench Verified(Verified subset (500 problems)) 80.84% · 附件证据(ev-op46-card-swe-verified)
Claude Opus 4.6 的 SWE-bench Verified · Verified subset (500 problems) 官方出处。Max effort · 80.84%。见《Claude Opus 4.6 System Card》PDF p.19, §2.4, SWE-bench Verified paragraph。证据 ev-op46-card-swe-verified。
原文与定位
Claude Opus 4.6 · SWE-bench Verified(Verified subset (500 problems)),Max effort配置,Pass rate 80.84%(具体成绩公布时间未确认)。见《Claude Opus 4.6 System Card》PDF p.19, §2.4, SWE-bench Verified paragraph。 查看成绩条件
Claude Opus 4.6 achieves 80.84% on SWE-bench Verified. Our SWE-bench results are averaged over 25 trials, each run with adaptive thinking, max effort, default sampling settings (temperature, top_p), and with the thinking blocks included in the sampling results.
- 官方材料
- Claude Opus 4.6 System Card · PDF 第 19 页
- 定位
- 附件 · PDF p.19, §2.4, SWE-bench Verified paragraph
- 材料版本
- Current PDF/TXT; local source 本地留存副本 and 本地留存副本; 213 pages; changelog through 2026-03-06; full text and visuals reviewed.
- 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 4.6 · SWE-bench Verified(Verified subset (500 problems)),Max effort配置,Pass rate 80.84%(具体成绩公布时间未确认)。官方出处 Claude Opus 4.6 System Card(附件,Current PDF/TXT; local source 本地留存副本 and 本地留存副本; 213 pages; changelog through 2026-03-06; full text and visuals reviewed.)PDF p.19, §2.4, SWE-bench Verified paragraph,https://www.anthropic.com/claude-opus-4-6-system-card#page=19。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-4-6/ev-op46-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 rate80.84%原文 80.84% · 单位 % 公布 未确认 | Max effort
| 直接支持该成绩 在排名中查看此成绩 |