Development

A benchmark record is beaten under thirty minutes later

development2026-02-06

Claim

Claude Opus 4.6 took the Terminal Bench 2.0 record at 65.4% and was beaten by GPT-5.3-Codex at 77.3% less than thirty minutes later.

Reported metric: 65.4% → 77.3%

Reporting date: (not necessarily the event date)

Sources

Attribution and reporting linked by this record; links do not imply independent verification.

The corpus records dates. This one needs a clock.

Knowledge graphopen in browser →

About
Reported in
Actor
Evidence for
Supersedes
Referenced by

Cite or correct this record

Suggest a correction — opens a GitHub draft for you to review and submit.