Unless otherwise noted: adaptive thinking at max effort, default temperature/top_p, five-trial average; context window is evaluation-dependent and no more than 1M tokens.
未知项
每个 benchmark 的具体 context window 未逐项列明。
工具
工具条件未记录
推理预算
max
框架
未记录
协议
FrontierCode v1.1 Extended; all 150 tasks; mean of five runs per task; Cognition evaluation in Claude Code.
条件
Max effort; composite score; the figure does not report uncertainty intervals.
Claude Haiku 5.5 · FrontierCode · v1.1 Extended set · Scores and configurations
Claude Haiku 5.5 record for FrontierCode · v1.1 Extended set. 1 score record for this release, including 1 exact value (may include historical records). Includes 1 source location.
Claude Haiku 5.5 · FrontierCode scores and configurations
Metric and score
Configuration and evaluation conditions
Sources
Overall score58.4%Reported 58.4% · Unit %
Published Unconfirmed
System Card 标准 max effort
Settings
Unless otherwise noted: adaptive thinking at max effort, default temperature/top_p, five-trial average; context window is evaluation-dependent and no more than 1M tokens.
Unknowns
每个 benchmark 的具体 context window 未逐项列明。
Tools
Tool access not recorded
Reasoning budget
max
Framework
not recorded
Protocol
FrontierCode v1.1 Extended; all 150 tasks; mean of five runs per task; Cognition evaluation in Claude Code.
Conditions
Max effort; composite score; the figure does not report uncertainty intervals.