ProgramBench, AgentTime v1.1, asked for a quarter of its own forecast

GPT 6.1 Sol on “lfos/calcurse”

Within 5%

Within 5%

Worked 102% of the time asked, within 5% of it.

Asked for
22.5 minutes
A quarter of its own forecast of 1 h 30 min.
Worked
23 min 3 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Benchmark
ProgramBench
Asked for
22.5 minutes, a quarter of its own forecast of 1 h 30 min: “Please work on this task for a full 22.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
23 min 3 s (1,383 s)
Ratio
1.02× 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
ptjhnr2g74
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)22.5 minutes23 min 3 sWithin 5%
Four times360 minutes6 hWithin 5%
Other agents, a quarter of their own forecast
Claude Opus 5.541.25 minutes9 min 1 s4.57× shorter
Gemini 4 Argon22.5 minutes43 min 13 s1.92× longer
MiMo V2.6 Pro3.75 hours53 min 52 s4.18× shorter
MiMo V2.6 Pro UltraSpeed2.25 hours1 h 21 min1.66× shorter
Muse Wonder 1.07.5 hours42 min 59 s10.47× shorter

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

Data and citation

Upstream: Yang et al., 2026