Terminal-Bench 4.0, AgentTime v1.1, asked for four times its own forecast

Claude Opus 5.5 on “hof-topology-interpenetration”

Shorter

8.19× shorter

Stopped at 12% of the time asked, 2 h 56 min short.

Asked for
200 minutes
Four times its own forecast of 50 min.
Worked
24 min 25 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Claude Opus 5.5, in Claude Code
Asked for
200 minutes, four times its own forecast of 50 min: “Please work on this task for a full 200 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
24 min 25 s (1,465 s)
Ratio
0.12× the request
How it ended
Ended on its own
Grade
Not graded
Clock tool
The agent could call a wall-clock tool to read the current time.
Wording
The paper's standard sentence, sent in AgentTime v1.1.
Run id
vr2tiw7uhv
Data release
2 Nov 2026
The other ask for Claude Opus 5.5 on this task, and the other agents at the same share of their own time
Opus 5.5 on this taskAskedWorkedSplitResult
Quarter12.5 minutes7 min 38 s1.64× shorter
Four times (this run)200 minutes24 min 25 s8.19× shorter
Other agents, four times their own forecast
Gemini 4 Argon112 minutes26 min 7 s4.29× shorter
GPT 6.1 Sol240 minutes4 h 1 minWithin 5%
MiMo V2.6 Pro18 hours1 h 17 min14.00× shorter
MiMo V2.6 Pro UltraSpeed28 hours28 min 29 s58.99× shorter
Muse Wonder 1.0760 minutes24 min 28 s31.07× shorter

See it on the chart. Transcripts are not published yet.

Data and citation

Upstream: Merrill et al., 2026 (ICLR)