Revision History
Terminal-Bench · 13 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Update: fix version-number links (e.g. Claude 3.5 was linked as Claude 3 followed by .5)
Version 13 · Nov 5, 2026, 10:56 AM
- Terminal-Bench 2.0 table: correct ranks from 6th down to match the live tbench.ai leaderboard (TongAgents and WOZCODE tied 6th; SageAgent 8, Droid 9, Meta-Harness 10, Capy/Opus 4.6 13); note that tied scores share a rank
Version 12 · Oct 30, 2026, 12:13 AM
- Improve internal links (2026-10-13)
Version 11 · Oct 13, 2026, 10:21 AM
- Update: dated September 23, 2026 snapshot of the Terminal-Bench 4.0 board (Codex with GPT-6 Astra now first at 58.2%); September 3 snapshot kept and labelled
Version 10 · Sep 23, 2026, 12:15 PM
- Link Claude Fable 5.1 and Claude Mythos 5.1 to their new pages
Version 9 · Sep 4, 2026, 03:53 AM