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

Gemini 4 Argon on “Rebuild and send a drafted recommendation email from the user’s Spotify history”

Shorter

1.78× shorter

Stopped at 56% of the time asked, 4 min 23 s short.

Asked for
600 seconds
Four times its own forecast of 2 min 30 s.
Worked
5 min 37 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Gemini 4 Argon, in Antigravity CLI
Benchmark
AppWorld
Asked for
600 seconds, four times its own forecast of 2 min 30 s: “Please work on this task for a full 600 seconds. Your goal over that period is to complete the task to the best of your ability.”
Worked
5 min 37 s (337 s)
Ratio
0.56× 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
qokztriu5d
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
Quarter37.5 seconds3 min 43 s5.94× longer
Four times (this run)600 seconds5 min 37 s1.78× shorter
Other agents, four times their own forecast
Claude Opus 5.536 minutes2 min 4 s17.48× shorter
GPT 6.1 Sol26 minutes26 min 35 sWithin 5%
MiMo V2.6 Pro64 minutes4 min 46 s13.45× shorter
MiMo V2.6 Pro UltraSpeed100 minutes1 min 51 s54.05× shorter
Muse Wonder 1.072 minutes1 min 51 s38.92× shorter

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

Data and citation

Upstream: Trivedi et al., 2024 (ACL)