Claude Sonnet 4.6 · BrowseComp(1,266 题;单代理与多代理按协议区分),No thinking, max effort配置,Accuracy 64.69%(具体成绩公布时间未确认)。见《Claude Sonnet 4.6 System Card》§2.20.1.2, PDF p.45。 查看成绩条件
Claude Sonnet 4.6 · BrowseComp(1,266 题;单代理与多代理按协议区分),No thinking, max effort配置,Accuracy 69.67%(具体成绩公布时间未确认)。见《Claude Sonnet 4.6 System Card》§2.20.1.2, PDF p.45。 查看成绩条件
Performance improves with test-time compute: Claude Sonnet 4.6 scored 64.69% when limited to 1M sampled tokens, 69.67% at 3M, and 74.01% at 10M.
PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10
版本公布
未确认
披露时间
· 后补披露
采集时间
历史依据
暂无可靠历史版本依据
引用文本
Claude Sonnet 4.6 · BrowseComp(1,266 题;单代理与多代理按协议区分),No thinking, max effort配置,Accuracy 64.69%(具体成绩公布时间未确认);Claude Sonnet 4.6 · BrowseComp(1,266 题;单代理与多代理按协议区分),No thinking, max effort配置,Accuracy 69.67%(具体成绩公布时间未确认)。官方出处 Claude Sonnet 4.6 System Card(附件,PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10)§2.20.1.2, PDF p.45,https://www.anthropic.com/claude-sonnet-4-6-system-card#page=45。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-4-6/ev46-browsecomp-scaling/
关联成绩与条件
Claude Sonnet 4.6 · BrowseComp 成绩与配置
指标与成绩
配置与评测条件
证据关系
Accuracy64.69%原文 64.69% · 单位 %
公布 未确认
No thinking, max effort
设置
Thinking disabled; max effort.
未知项
Sampling settings are not specified in every evaluation.
工具
允许工具
推理预算
max
框架
未记录
协议
BrowseComp single-agent with 1M sampled-token limit.
条件
Context compaction; the model is told the total-token limit. Web search, web fetch and programmatic tool calling; no thinking.
Official source for BrowseComp · 1,266 题;单代理与多代理按协议区分 in Claude Sonnet 4.6. See “Claude Sonnet 4.6 System Card” at §2.20.1.2, PDF p.45. Evidence ev46-browsecomp-scaling.
Claude Sonnet 4.6 · BrowseComp (1,266 题;单代理与多代理按协议区分), configuration No thinking, max effort, Accuracy 64.69% (exact score publication date unconfirmed). See “Claude Sonnet 4.6 System Card” at §2.20.1.2, PDF p.45. View score conditions
Claude Sonnet 4.6 · BrowseComp (1,266 题;单代理与多代理按协议区分), configuration No thinking, max effort, Accuracy 69.67% (exact score publication date unconfirmed). See “Claude Sonnet 4.6 System Card” at §2.20.1.2, PDF p.45. View score conditions
Performance improves with test-time compute: Claude Sonnet 4.6 scored 64.69% when limited to 1M sampled tokens, 69.67% at 3M, and 74.01% at 10M.
PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10
Version published
Unconfirmed
Disclosure date
· Later disclosure
Collected
Historical basis
No reliable historical version evidence available
Citation text
Claude Sonnet 4.6 · BrowseComp (1,266 题;单代理与多代理按协议区分), configuration No thinking, max effort, Accuracy 64.69% (exact score publication date unconfirmed); Claude Sonnet 4.6 · BrowseComp (1,266 题;单代理与多代理按协议区分), configuration No thinking, max effort, Accuracy 69.67% (exact score publication date unconfirmed). Official source: Claude Sonnet 4.6 System Card (Attachment, PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10), §2.20.1.2, PDF p.45, https://www.anthropic.com/claude-sonnet-4-6-system-card#page=45. Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-4-6/ev46-browsecomp-scaling/?lang=en
Related scores and conditions
Claude Sonnet 4.6 · BrowseComp scores and configurations
Metric and score
Configuration and evaluation conditions
Evidence relation
Accuracy64.69%Reported 64.69% · Unit %
Published Unconfirmed
No thinking, max effort
Settings
Thinking disabled; max effort.
Unknowns
Sampling settings are not specified in every evaluation.
Tools
Tools allowed
Reasoning budget
max
Framework
not recorded
Protocol
BrowseComp single-agent with 1M sampled-token limit.
Conditions
Context compaction; the model is told the total-token limit. Web search, web fetch and programmatic tool calling; no thinking.