Standard configuration: adaptive thinking, max effort
设置
Adaptive thinking at max effort; default temperature and top_p; five-trial averages unless otherwise stated.
未知项
Context-window size is evaluation-dependent; standard configuration is 1M tokens.
工具
允许工具
推理预算
max
框架
未记录
协议
OfficeQA exact-match accuracy; mean of five trials.
条件
Agentic evaluation with extracted-text documents and code execution. Episodes reaching per-turn output limit before a final answer are incorrect; about 9–15% do so. Raw-PDF harnesses yield lower scores and are not directly comparable.
Differences are calculated only for unique, active, exact values between direct predecessors when the unit, metric, and explicitly recorded configuration and conditions match.
· 1 score record for this release, including 1 exact value (may include historical records)
Records for the direct predecessor do not provide matching conditions for a valid difference.
Claude Sonnet 5 · OfficeQA scores and configurations
Metric and score
Configuration and evaluation conditions
Sources
Exact-match accuracy73.3%Reported 73.3% · Unit %
Published Unconfirmed
Standard configuration: adaptive thinking, max effort
Settings
Adaptive thinking at max effort; default temperature and top_p; five-trial averages unless otherwise stated.
Unknowns
Context-window size is evaluation-dependent; standard configuration is 1M tokens.
Tools
Tools allowed
Reasoning budget
max
Framework
not recorded
Protocol
OfficeQA exact-match accuracy; mean of five trials.
Conditions
Agentic evaluation with extracted-text documents and code execution. Episodes reaching per-turn output limit before a final answer are incorrect; about 9–15% do so. Raw-PDF harnesses yield lower scores and are not directly comparable.