Claude Opus 5 · DeepSWE(v1.1) · 附件证据(ev-op5-card-deepswe-sweep)
Claude Opus 5 的 DeepSWE · v1.1 官方出处。见《Claude Opus 5 System Card》System Card PDF p.154, Figure 8.3.A; Opus 5 series labels。证据 ev-op5-card-deepswe-sweep。
Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Max effort配置,Score 68.8%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.154, Figure 8.3.A; Opus 5 series labels。 查看成绩条件
Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Low effort配置,Score 57.7%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.154, Figure 8.3.A; Opus 5 series labels。 查看成绩条件
Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Medium effort配置,Score 66.9%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.154, Figure 8.3.A; Opus 5 series labels。 查看成绩条件
Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · High effort配置,Score 68%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.154, Figure 8.3.A; Opus 5 series labels。 查看成绩条件
Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Xhigh effort配置,Score 69.7%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.154, Figure 8.3.A; Opus 5 series labels。 查看成绩条件
视觉转录(Claude Opus 5 图中明确标注):low 57.7%; medium 66.9%; high 68.0%; xhigh 69.7%; max 68.8%.
附件 · System Card PDF p.154, Figure 8.3.A; Opus 5 series labels
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 · DeepSWE(v1.1),DeepSWE v1.1 · Max effort配置,Score 68.8%(具体成绩公布时间未确认);Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Low effort配置,Score 57.7%(具体成绩公布时间未确认);Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Medium effort配置,Score 66.9%(具体成绩公布时间未确认);Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · High effort配置,Score 68%(具体成绩公布时间未确认);Claude Opus 5 · DeepSWE(v1.1),DeepSWE v1.1 · Xhigh effort配置,Score 69.7%(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.154, Figure 8.3.A; Opus 5 series labels,https://www.anthropic.com/claude-opus-5-system-card#page=154。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-deepswe-sweep/
关联成绩与条件
Claude Opus 5 · DeepSWE 成绩与配置
指标与成绩
配置与评测条件
证据关系
Score68.8%原文 68.8% · 单位 %
公布 未确认
DeepSWE v1.1 · Max effort
设置
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: max.
未知项
The card does not state any additional sampling parameters beyond the cited evaluation description.
工具
工具条件未记录
推理预算
max
框架
未记录
协议
DeepSWE v1.1; §8.3 prose reports the max-effort summary score; Figure 8.3.A max point is 68.8%.
条件
68.8%, average over five trials. Keep distinct effort-sweep points as separate results.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: low.
未知项
The card does not state any additional sampling parameters beyond the cited evaluation description.
工具
工具条件未记录
推理预算
low
框架
未记录
协议
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
条件
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: medium.
未知项
The card does not state any additional sampling parameters beyond the cited evaluation description.
工具
工具条件未记录
推理预算
medium
框架
未记录
协议
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
条件
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: high.
未知项
The card does not state any additional sampling parameters beyond the cited evaluation description.
工具
工具条件未记录
推理预算
high
框架
未记录
协议
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
条件
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: xhigh.
未知项
The card does not state any additional sampling parameters beyond the cited evaluation description.
工具
工具条件未记录
推理预算
xhigh
框架
未记录
协议
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
条件
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
Claude Opus 5 · DeepSWE (v1.1) · Attachment evidence (ev-op5-card-deepswe-sweep)
Official source for DeepSWE · v1.1 in Claude Opus 5. See “Claude Opus 5 System Card” at System Card PDF p.154, Figure 8.3.A; Opus 5 series labels. Evidence ev-op5-card-deepswe-sweep.
Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Max effort, Score 68.8% (exact score publication date unconfirmed). See “Claude Opus 5 System Card” at System Card PDF p.154, Figure 8.3.A; Opus 5 series labels. View score conditions
Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Low effort, Score 57.7% (exact score publication date unconfirmed). See “Claude Opus 5 System Card” at System Card PDF p.154, Figure 8.3.A; Opus 5 series labels. View score conditions
Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Medium effort, Score 66.9% (exact score publication date unconfirmed). See “Claude Opus 5 System Card” at System Card PDF p.154, Figure 8.3.A; Opus 5 series labels. View score conditions
Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · High effort, Score 68% (exact score publication date unconfirmed). See “Claude Opus 5 System Card” at System Card PDF p.154, Figure 8.3.A; Opus 5 series labels. View score conditions
Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Xhigh effort, Score 69.7% (exact score publication date unconfirmed). See “Claude Opus 5 System Card” at System Card PDF p.154, Figure 8.3.A; Opus 5 series labels. View score conditions
视觉转录(Claude Opus 5 图中明确标注):low 57.7%; medium 66.9%; high 68.0%; xhigh 69.7%; max 68.8%.
Attachment · System Card PDF p.154, Figure 8.3.A; Opus 5 series labels
Material version
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.
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 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Max effort, Score 68.8% (exact score publication date unconfirmed); Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Low effort, Score 57.7% (exact score publication date unconfirmed); Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Medium effort, Score 66.9% (exact score publication date unconfirmed); Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · High effort, Score 68% (exact score publication date unconfirmed); Claude Opus 5 · DeepSWE (v1.1), configuration DeepSWE v1.1 · Xhigh effort, Score 69.7% (exact score publication date unconfirmed). Official source: Claude Opus 5 System Card (Attachment, 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.154, Figure 8.3.A; Opus 5 series labels, https://www.anthropic.com/claude-opus-5-system-card#page=154. Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-deepswe-sweep/?lang=en
Related scores and conditions
Claude Opus 5 · DeepSWE scores and configurations
Metric and score
Configuration and evaluation conditions
Evidence relation
Score68.8%Reported 68.8% · Unit %
Published Unconfirmed
DeepSWE v1.1 · Max effort
Settings
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: max.
Unknowns
The card does not state any additional sampling parameters beyond the cited evaluation description.
Tools
Tool access not recorded
Reasoning budget
max
Framework
not recorded
Protocol
DeepSWE v1.1; §8.3 prose reports the max-effort summary score; Figure 8.3.A max point is 68.8%.
Conditions
68.8%, average over five trials. Keep distinct effort-sweep points as separate results.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: low.
Unknowns
The card does not state any additional sampling parameters beyond the cited evaluation description.
Tools
Tool access not recorded
Reasoning budget
low
Framework
not recorded
Protocol
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
Conditions
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: medium.
Unknowns
The card does not state any additional sampling parameters beyond the cited evaluation description.
Tools
Tool access not recorded
Reasoning budget
medium
Framework
not recorded
Protocol
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
Conditions
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: high.
Unknowns
The card does not state any additional sampling parameters beyond the cited evaluation description.
Tools
Tool access not recorded
Reasoning budget
high
Framework
not recorded
Protocol
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
Conditions
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.
113 long-horizon software-engineering tasks; Figure 8.3.A effort/cost sweep. The figure does not state effort-specific trial counts. Explicit adaptive-thinking effort: xhigh.
Unknowns
The card does not state any additional sampling parameters beyond the cited evaluation description.
Tools
Tool access not recorded
Reasoning budget
xhigh
Framework
not recorded
Protocol
DeepSWE v1.1; Figure 8.3.A effort/cost sweep.
Conditions
113 long-horizon coding tasks; explicit figure-labeled effort; figure-specific run count not stated. Keep the chart series distinct from the prose/table max value, which is also 68.8%.