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

Gemini 4 Argon on “stranger6667/jsonschema”

Longer

1.37× longer

Worked 37% past the request, 8 min 21 s over.

Asked for
22.5 minutes
A quarter of its own forecast of 1 h 30 min.
Worked
30 min 51 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
22.5 minutes, a quarter of its own forecast of 1 h 30 min: “Please work on this task for a full 22.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
30 min 51 s (1,851 s)
Ratio
1.37× 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
b5of4gbbcz
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
Quarter (this run)22.5 minutes30 min 51 s1.37× longer
Four times360 minutes1 h 48 min3.34× shorter
Other agents, a quarter of their own forecast
Claude Opus 5.542.5 minutes33 min 31 s1.27× shorter
GPT 6.1 Sol15 minutes15 min 26 sWithin 5%
MiMo V2.6 Pro3.25 hours1 h 4 min3.05× shorter
MiMo V2.6 Pro UltraSpeed1.75 hours50 min 4 s2.10× shorter
Muse Wonder 1.02.75 hours16 min 47 s9.83× shorter

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

Data and citation

Upstream: Yang et al., 2026