0

Terminal Bench 2 Verified

Verified tasks from terminal bench 2, created by z.ai

Domain
rl-env
License
unknown
Published
Feb 2026

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
OpenReward
Attribution policy →

Top score 61.8 by Claude Opus 4.5 - 3 models reporting (2 frontier)

Score history

3
0255075100Sep 25Nov 25Jan 26Claude Sonnet 4.5Claude Opus 4.5

Top models

3
Terminal Bench 2 VerifiedBar chart with 3 bars. Highest value: Claude Opus 4.5 at 61.8.
3 models

FAQ

What is Terminal Bench 2 Verified?
Verified tasks from terminal bench 2, created by z.ai
What is the current top score on Terminal Bench 2 Verified?
The top reported score is 61.8 by Claude Opus 4.5, across 3 models reporting (2 from frontier labs).
What license is Terminal Bench 2 Verified under?
Terminal Bench 2 Verified is available under unknown.