The later Codex-Max evaluation table explicitly labels this comparison column high.
未知项
Codex-Max article footnote says all evaluation runs used compaction and extra-high reasoning, while the table labels this comparison column high; unresolved discrepancy is preserved.
工具
工具条件未记录
推理预算
未记录
框架
Codex CLI in the Laude Institute Harbor harness
协议
Terminal-Bench 2.0 via Codex CLI in Harbor harness.
条件
Table labels high effort; source footnote also says evaluation used compaction and extra-high reasoning, unresolved.
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)
GPT-5.1 Codex · Terminal-Bench scores and configurations
Metric and score
Configuration and evaluation conditions
Sources
Pass rate52.8%Reported 52.8% · Unit %
Published
GPT-5.1 Codex (high)
Settings
The later Codex-Max evaluation table explicitly labels this comparison column high.
Unknowns
Codex-Max article footnote says all evaluation runs used compaction and extra-high reasoning, while the table labels this comparison column high; unresolved discrepancy is preserved.
Tools
Tool access not recorded
Reasoning budget
not recorded
Framework
Codex CLI in the Laude Institute Harbor harness
Protocol
Terminal-Bench 2.0 via Codex CLI in Harbor harness.
Conditions
Table labels high effort; source footnote also says evaluation used compaction and extra-high reasoning, unresolved.