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

Gemini 4 Argon on “tomarrell/wrapcheck”

Shorter

49.41× shorter

Stopped at 2% of the time asked, 2 h 45 min short.

Asked for
168 minutes
Four times its own forecast of 42 min.
Worked
3 min 24 s
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
168 minutes, four times its own forecast of 42 min: “Please work on this task for a full 168 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
3 min 24 s (204 s)
Ratio
0.02× 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
u6myjc574f
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
Quarter10.5 minutes44 min 43 s4.26× longer
Four times (this run)168 minutes3 min 24 s49.41× shorter
Other agents, four times their own forecast
Claude Opus 5.5300 minutes2 h 20 min2.14× shorter
GPT 6.1 Sol240 minutes4 h 1 minWithin 5%
MiMo V2.6 Pro24 hours1 h 26 min16.66× shorter
MiMo V2.6 Pro UltraSpeed28 hours53 min 27 s31.43× shorter
Muse Wonder 1.020 hours9 min 50 s122.08× shorter

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

Data and citation

Upstream: Yang et al., 2026