Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark) · 附件证据(ev55-visual-draco-effort)

Claude Sonnet 5.5 的 DRACO · 100-task Perplexity deep-research benchmark 官方出处。见《Claude Sonnet 5.5 System Card》§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription。证据 ev55-visual-draco-effort。

原文与定位

Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Low effort配置,Normalized score 64.6%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Medium effort配置,Normalized score 71.2%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),High effort配置,Normalized score 81%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Xhigh effort配置,Normalized score 85.7%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Max effort配置,Normalized score 87%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription。 查看成绩条件

Claude Sonnet 5.5: Low 64.6%; Med 71.2%; High 81.0%; Xhigh 85.7%; Max 87.0%.
官方材料
Claude Sonnet 5.5 System Card · PDF 第 121 页
定位
附件 · §8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription
材料版本
PDF cover dated 2026-09-28; 148 pages; collected 2026-11-10
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 Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Low effort配置,Normalized score 64.6%(具体成绩公布时间未确认);Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Medium effort配置,Normalized score 71.2%(具体成绩公布时间未确认);Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),High effort配置,Normalized score 81%(具体成绩公布时间未确认);Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Xhigh effort配置,Normalized score 85.7%(具体成绩公布时间未确认);Claude Sonnet 5.5 · DRACO(100-task Perplexity deep-research benchmark),Max effort配置,Normalized score 87%(具体成绩公布时间未确认)。官方出处 Claude Sonnet 5.5 System Card(附件,PDF cover dated 2026-09-28; 148 pages; collected 2026-11-10)§8.11.2, Figure 8.11.2.A, PDF p.121; chart-label transcription,https://www.anthropic.com/claude-sonnet-5-5-system-card#page=121。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-5-5/ev55-visual-draco-effort/

关联成绩与条件

Claude Sonnet 5.5 · DRACO 成绩与配置
指标与成绩配置与评测条件证据关系
Normalized score64.6%原文 64.6% · 单位 %
公布 未确认
Low effort
设置
Low reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
允许工具
推理预算
low
框架
未记录
协议
DRACO, 100 curated tasks; normalized score from five independent rubric-grading runs per response.
条件
Figure labels the effort Low; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; Claude Opus 4.6 grader. The report says its modified tools, prompts, and grader make results not directly comparable to Perplexity's version.

直接支持该成绩

在排名中查看此成绩
Normalized score71.2%原文 71.2% · 单位 %
公布 未确认
Medium effort
设置
Medium reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
允许工具
推理预算
med
框架
未记录
协议
DRACO, 100 curated tasks; normalized score from five independent rubric-grading runs per response.
条件
Figure labels the effort Med; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; Claude Opus 4.6 grader. The report says its modified tools, prompts, and grader make results not directly comparable to Perplexity's version.

直接支持该成绩

在排名中查看此成绩
Normalized score81%原文 81.0% · 单位 %
公布 未确认
High effort
设置
High reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
允许工具
推理预算
high
框架
未记录
协议
DRACO, 100 curated tasks; normalized score from five independent rubric-grading runs per response.
条件
Figure labels the effort High; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; Claude Opus 4.6 grader. The report says its modified tools, prompts, and grader make results not directly comparable to Perplexity's version.

直接支持该成绩

在排名中查看此成绩
Normalized score85.7%原文 85.7% · 单位 %
公布 未确认
Xhigh effort
设置
Xhigh reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
允许工具
推理预算
xhigh
框架
未记录
协议
DRACO, 100 curated tasks; normalized score from five independent rubric-grading runs per response.
条件
Figure labels the effort Xhigh; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; Claude Opus 4.6 grader. The report says its modified tools, prompts, and grader make results not directly comparable to Perplexity's version.

直接支持该成绩

在排名中查看此成绩
Normalized score87%原文 87.0% · 单位 %
公布 未确认
Max effort
设置
Max reasoning effort.
未知项
Adaptive thinking status is benchmark-specific.
工具
允许工具
推理预算
max
框架
未记录
协议
DRACO, 100 curated tasks; normalized score from five independent rubric-grading runs per response.
条件
Figure labels the effort Max; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; Claude Opus 4.6 grader. The report says its modified tools, prompts, and grader make results not directly comparable to Perplexity's version.

直接支持该成绩

在排名中查看此成绩