TUA-Bench, AgentTime v1.1, asked for a quarter of its own forecast

GPT 6.1 Sol on “Author homepage bookmarks”

Longer

1.13× longer

Worked 13% past the request, 15 s over.

Asked for
2 minutes
A quarter of its own forecast of 8 min.
Worked
2 min 15 s
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
2 minutes, a quarter of its own forecast of 8 min: “Please work on this task for a full 2 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
2 min 15 s (135 s)
Ratio
1.13× 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
4ua3hikeif
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
Quarter (this run)2 minutes2 min 15 s1.13× longer
Four times32 minutes32 min 22 sWithin 5%
Other agents, a quarter of their own forecast
Claude Opus 5.51.5 minutes1 min 9 s1.30× shorter
Gemini 4 Argon37.5 seconds4 min 4 s6.50× longer
MiMo V2.6 Pro6 minutes12 min 8 s2.02× longer
MiMo V2.6 Pro UltraSpeed7.5 minutes7 min 40 sWithin 5%
Muse Wonder 1.05.5 minutes5 min 11 s1.06× shorter

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

Data and citation

Upstream: Chen et al., 2026