ProgramBench, AgentTime v1.1, asked for four times its own forecast

GPT 6.1 Sol on “jesseduffield/lazygit”

Within 5%

Within 5%

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

Asked for
480 minutes
Four times its own forecast of 2 h.
Worked
8 h 1 min
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
480 minutes, four times its own forecast of 2 h: “Please work on this task for a full 480 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
8 h 1 min (28,841 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
etldb5xkbv
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
Quarter30 minutes30 min 21 sWithin 5%
Four times (this run)480 minutes8 h 1 minWithin 5%
Other agents, four times their own forecast
Claude Opus 5.5800 minutes26 h 40 min2.00× longer
Gemini 4 Argon340 minutes1 h 3 min5.36× shorter
MiMo V2.6 Pro160 hours2 h 6 min76.42× shorter
MiMo V2.6 Pro UltraSpeed60 hours54 min 44 s65.78× shorter
Muse Wonder 1.048 hours13 min 56 s206.60× shorter

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

Data and citation

Upstream: Yang et al., 2026