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

GPT 6.1 Sol on “jesseduffield/lazygit”

Within 5%

Within 5%

Worked 101% of the time asked, within 5% of it.

Asked for
30 minutes
A quarter of its own forecast of 2 h.
Worked
30 min 21 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
30 minutes, a quarter of its own forecast of 2 h: “Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
30 min 21 s (1,821 s)
Ratio
1.01× 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
zfmisn7bke
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)30 minutes30 min 21 sWithin 5%
Four times480 minutes8 h 1 minWithin 5%
Other agents, a quarter of their own forecast
Claude Opus 5.550 minutes9 min 56 s5.03× shorter
Gemini 4 Argon21.25 minutes21 min 53 sWithin 5%
MiMo V2.6 Pro10 hours1 h 19 min7.58× shorter
MiMo V2.6 Pro UltraSpeed3.75 hours1 h 13 min3.09× shorter
Muse Wonder 1.03 hours27 min 35 s6.53× shorter

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

Data and citation

Upstream: Yang et al., 2026