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

GPT 6.1 Sol on “Rebuild a retired production risk scorer from black-box behavior”

Within 5%

Within 5%

Worked 100% of the time asked, within 5% of it.

Asked for
200 minutes
Four times its own forecast of 50 min.
Worked
3 h 21 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
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
3 h 21 min (12,041 s)
Ratio
1.00× 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
gbsaaqo3d7
Data release
2 Nov 2026
The other ask for GPT 6.1 Sol on this task, and the other agents at the same share of their own time
Sol 6.1 on this taskAskedWorkedSplitResult
Quarter12.5 minutes12 min 41 sWithin 5%
Four times (this run)200 minutes3 h 21 minWithin 5%
Other agents, four times their own forecast
Claude Opus 5.5220 minutes51 min 46 s4.25× shorter
Gemini 4 Argon80 minutes21 min 54 s3.65× shorter
MiMo V2.6 Pro480 minutes2 h 9 min3.72× shorter
MiMo V2.6 Pro UltraSpeed600 minutes34 min 58 s17.16× shorter
Muse Wonder 1.0360 minutes13 min 14 s27.21× shorter

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)