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

GPT 6.1 Sol on “Optimize cold plate”

Within 5%

Within 5%

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

Asked for
280 minutes
Four times its own forecast of 1 h 10 min.
Worked
4 h 41 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Benchmark
TUA-Bench
Asked for
280 minutes, four times its own forecast of 1 h 10 min: “Please work on this task for a full 280 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
4 h 41 min (16,871 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
775o7f4vdi
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
Quarter17.5 minutes17 min 57 sWithin 5%
Four times (this run)280 minutes4 h 41 minWithin 5%
Other agents, four times their own forecast
Claude Opus 5.5360 minutes2 h 10 min2.78× shorter
Gemini 4 Argon112 minutes1 h 33 min1.20× shorter
MiMo V2.6 Pro32 hours1 h 10 min27.61× shorter
MiMo V2.6 Pro UltraSpeed660 minutes44 min 5 s14.97× shorter
Muse Wonder 1.018 hours8 min 9 s132.52× shorter

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

Data and citation

Upstream: Chen et al., 2026