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

Gemini 4 Argon on “stranger6667/jsonschema”

Shorter

3.34× shorter

Stopped at 30% of the time asked, 4 h 12 min short.

Asked for
360 minutes
Four times its own forecast of 1 h 30 min.
Worked
1 h 48 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
360 minutes, four times its own forecast of 1 h 30 min: “Please work on this task for a full 360 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
1 h 48 min (6,470 s)
Ratio
0.30× 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
nsv4yywlp3
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
Quarter22.5 minutes30 min 51 s1.37× longer
Four times (this run)360 minutes1 h 48 min3.34× shorter
Other agents, four times their own forecast
Claude Opus 5.5680 minutes10 h 56 minWithin 5%
GPT 6.1 Sol240 minutes4 hWithin 5%
MiMo V2.6 Pro52 hours1 h 56 min26.83× shorter
MiMo V2.6 Pro UltraSpeed28 hours1 h 15 min22.35× shorter
Muse Wonder 1.044 hours16 min 31 s159.90× shorter

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

Data and citation

Upstream: Yang et al., 2026