Claude Opus 5.5 · FrontierCode(v1.1 Main set),Adaptive Max配置,Overall score 54.4%(具体成绩公布时间未确认)。见《Claude Opus 5.5 System Card》§8.4, pp.176-177, Figures 8.4.A-B。 查看成绩条件
Claude Opus 5.5 · FrontierCode(v1.1 Main set),FrontierCode v1.1 Main medium effort配置,Overall score 54.6%(具体成绩公布时间未确认)。见《Claude Opus 5.5 System Card》§8.4, pp.176-177, Figures 8.4.A-B。 查看成绩条件
Opus 5.5 ranks first on FrontierCode (Main) with a 54.6% score ... Opus 5.5 also ranks first on FrontierCode (Extended) with a 65.3% score ... highest ... at medium effort ... At max effort, Opus 5.5 scores 54.4% on Main and 63.6% on Extended.
官方当前 System Card;封面日期 2026-09-22;本地 PDF/TXT:本地留存副本、current-opus-5-5-system-card.txt;230 页全文核读;当前修订时间未知
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 5.5 · FrontierCode(v1.1 Main set),Adaptive Max配置,Overall score 54.4%(具体成绩公布时间未确认);Claude Opus 5.5 · FrontierCode(v1.1 Main set),FrontierCode v1.1 Main medium effort配置,Overall score 54.6%(具体成绩公布时间未确认)。官方出处 Claude Opus 5.5 System Card(附件,官方当前 System Card;封面日期 2026-09-22;本地 PDF/TXT:本地留存副本、current-opus-5-5-system-card.txt;230 页全文核读;当前修订时间未知)§8.4, pp.176-177, Figures 8.4.A-B,https://www.anthropic.com/claude-opus-5-5-system-card#page=176。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5-5/ev-op55-card-frontiercode/
除特别注明外 adaptive thinking/max effort。 Source-wide conditions: Cognition ran the evaluations in Claude Code; mean@5. Main and Extended are separate benchmark definitions.
Explicit medium effort; Cognition evaluation in Claude Code. Mean@5.
未知项
No API-equivalent harness or run date is stated.
工具
工具条件未记录
推理预算
medium
框架
Cognition Claude Code harness
协议
FrontierCode v1.1 Main; Cognition Claude Code harness; mean@5; medium effort.
条件
Best reported effort. Evaluation has internet access; no timeout information is stated. Do not infer effort from output tokens. Source-wide conditions: Cognition ran the evaluations in Claude Code; mean@5. Main and Extended are separate benchmark definitions.
Claude Opus 5.5 · FrontierCode (v1.1 Main set) · Attachment evidence (ev-op55-card-frontiercode)
Official source for FrontierCode · v1.1 Main set in Claude Opus 5.5. See “Claude Opus 5.5 System Card” at §8.4, pp.176-177, Figures 8.4.A-B. Evidence ev-op55-card-frontiercode.
Claude Opus 5.5 · FrontierCode (v1.1 Main set), configuration Adaptive Max, Overall score 54.4% (exact score publication date unconfirmed). See “Claude Opus 5.5 System Card” at §8.4, pp.176-177, Figures 8.4.A-B. View score conditions
Claude Opus 5.5 · FrontierCode (v1.1 Main set), configuration FrontierCode v1.1 Main medium effort, Overall score 54.6% (exact score publication date unconfirmed). See “Claude Opus 5.5 System Card” at §8.4, pp.176-177, Figures 8.4.A-B. View score conditions
Opus 5.5 ranks first on FrontierCode (Main) with a 54.6% score ... Opus 5.5 also ranks first on FrontierCode (Extended) with a 65.3% score ... highest ... at medium effort ... At max effort, Opus 5.5 scores 54.4% on Main and 63.6% on Extended.
官方当前 System Card;封面日期 2026-09-22;本地 PDF/TXT:本地留存副本、current-opus-5-5-system-card.txt;230 页全文核读;当前修订时间未知
Version published
Unconfirmed
Disclosure date
Unconfirmed · Launch or later disclosure unconfirmed
Collected
Historical basis
No reliable historical version evidence available
Citation text
Claude Opus 5.5 · FrontierCode (v1.1 Main set), configuration Adaptive Max, Overall score 54.4% (exact score publication date unconfirmed); Claude Opus 5.5 · FrontierCode (v1.1 Main set), configuration FrontierCode v1.1 Main medium effort, Overall score 54.6% (exact score publication date unconfirmed). Official source: Claude Opus 5.5 System Card (Attachment, 官方当前 System Card;封面日期 2026-09-22;本地 PDF/TXT:本地留存副本、current-opus-5-5-system-card.txt;230 页全文核读;当前修订时间未知), §8.4, pp.176-177, Figures 8.4.A-B, https://www.anthropic.com/claude-opus-5-5-system-card#page=176. Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5-5/ev-op55-card-frontiercode/?lang=en
Related scores and conditions
Claude Opus 5.5 · FrontierCode scores and configurations
除特别注明外 adaptive thinking/max effort。 Source-wide conditions: Cognition ran the evaluations in Claude Code; mean@5. Main and Extended are separate benchmark definitions.
Explicit medium effort; Cognition evaluation in Claude Code. Mean@5.
Unknowns
No API-equivalent harness or run date is stated.
Tools
Tool access not recorded
Reasoning budget
medium
Framework
Cognition Claude Code harness
Protocol
FrontierCode v1.1 Main; Cognition Claude Code harness; mean@5; medium effort.
Conditions
Best reported effort. Evaluation has internet access; no timeout information is stated. Do not infer effort from output tokens. Source-wide conditions: Cognition ran the evaluations in Claude Code; mean@5. Main and Extended are separate benchmark definitions.