Adaptive thinking · max effort (card-wide default)
设置
System Card §8.1 default: adaptive thinking at max effort, default sampling settings (temperature/top_p), averaged over five trials unless otherwise noted; context varies by evaluation and does not exceed 1M tokens.
未知项
Per-benchmark exceptions and harness details are recorded on the relevant configurations/results.
工具
允许工具
推理预算
max
框架
未记录
协议
GDPval-AA v2; independent Artificial Analysis evaluation; 220 tasks across 44 occupations and 9 industries.
条件
1861 Elo at max effort; shell and web-browsing agent loop; retain the existing main-page evidence.
Claude Opus 5 · GDPval-AA · v2 · Scores and configurations
Claude Opus 5 record for GDPval-AA · v2. 2 score records for this release, including 2 exact values (may include historical records). Includes 2 source locations.
Claude Opus 5 · GDPval-AA scores and configurations
Metric and score
Configuration and evaluation conditions
Sources
Elo rating1861 EloReported 1861 Elo · Unit Elo
Published Unconfirmed
Adaptive thinking · max effort (card-wide default)
Settings
System Card §8.1 default: adaptive thinking at max effort, default sampling settings (temperature/top_p), averaged over five trials unless otherwise noted; context varies by evaluation and does not exceed 1M tokens.
Unknowns
Per-benchmark exceptions and harness details are recorded on the relevant configurations/results.
Tools
Tools allowed
Reasoning budget
max
Framework
not recorded
Protocol
GDPval-AA v2; independent Artificial Analysis evaluation; 220 tasks across 44 occupations and 9 industries.
Conditions
1861 Elo at max effort; shell and web-browsing agent loop; retain the existing main-page evidence.