除章节另有说明外,adaptive thinking at max effort,默认 temperature/top_p,平均 5 trials;评测上下文依任务而定且不超过 1M tokens。
未知项
未另记
工具
工具条件未记录
推理预算
max
框架
未记录
协议
Finance Agent Benchmark v2; Vals AI; adaptive thinking and max effort.
条件
Keep distinct from Finance Agent v1.1; v2 was released in May 2026 and is substantially harder. The announcement’s rounded 53.9% evidence remains attached.
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 Opus 4.8 · Finance Agent scores and configurations
Metric and score
Configuration and evaluation conditions
Sources
Accuracy53.92%Reported 53.92% · Unit %
Published Unconfirmed
System Card 默认配置
Settings
除章节另有说明外,adaptive thinking at max effort,默认 temperature/top_p,平均 5 trials;评测上下文依任务而定且不超过 1M tokens。
Unknowns
none recorded
Tools
Tool access not recorded
Reasoning budget
max
Framework
not recorded
Protocol
Finance Agent Benchmark v2; Vals AI; adaptive thinking and max effort.
Conditions
Keep distinct from Finance Agent v1.1; v2 was released in May 2026 and is substantially harder. The announcement’s rounded 53.9% evidence remains attached.