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

Gemini 4 Argon on “rvben/rumdl”

Shorter

9.16× shorter

Stopped at 11% of the time asked, 6 h 50 min short.

Asked for
460 minutes
Four times its own forecast of 1 h 55 min.
Worked
50 min 13 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
460 minutes, four times its own forecast of 1 h 55 min: “Please work on this task for a full 460 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
50 min 13 s (3,013 s)
Ratio
0.11× 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
wugatwltfm
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
Quarter28.75 minutes46 min 33 s1.62× longer
Four times (this run)460 minutes50 min 13 s9.16× shorter
Other agents, four times their own forecast
Claude Opus 5.516 hours15 h 51 minWithin 5%
GPT 6.1 Sol480 minutes8 hWithin 5%
MiMo V2.6 Pro80 hours3 h 6 min25.76× shorter
MiMo V2.6 Pro UltraSpeed100 hours1 h 54 min52.43× shorter
Muse Wonder 1.064 hours15 min 27 s248.54× shorter

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

Data and citation

Upstream: Yang et al., 2026