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

Gemini 4 Argon on “doxygen/doxygen”

Shorter

2.98× shorter

Stopped at 34% of the time asked, 4 h 39 min short.

Asked for
420 minutes
Four times its own forecast of 1 h 45 min.
Worked
2 h 21 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Gemini 4 Argon, in Antigravity CLI
Benchmark
ProgramBench
Asked for
420 minutes, four times its own forecast of 1 h 45 min: “Please work on this task for a full 420 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
2 h 21 min (8,452 s)
Ratio
0.34× 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
ak6g6t2ppg
Data release
2 Nov 2026
The other ask for Gemini 4 Argon on this task, and the other agents at the same share of their own time
Argon on this taskAskedWorkedSplitResult
Quarter26.25 minutes28 min 31 s1.09× longer
Four times (this run)420 minutes2 h 21 min2.98× shorter
Other agents, four times their own forecast
Claude Opus 5.516 hours14 h 54 min1.07× shorter
GPT 6.1 Sol420 minutes7 hWithin 5%
MiMo V2.6 Pro200 hours1 h 10 min172.34× shorter
MiMo V2.6 Pro UltraSpeed60 hours1 h 14 min48.89× shorter
Muse Wonder 1.072 hours2 h 12 min32.63× shorter

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

Data and citation

Upstream: Yang et al., 2026