CORE-Bench v1.1, AgentTime v1.1, asked for four times its own forecast

Gemini 4 Argon on “Reproduce a 20 mm transducer-array characterization”

Longer

2.00× longer

Worked 100% past the request, 14 min 1 s over.

Asked for
840 seconds
Four times its own forecast of 3 min 30 s.
Worked
28 min 1 s
Grade
Not graded
How it ended
Stopped at AgentTime's time limit
Every detail of this run
Agent
Gemini 4 Argon, in Antigravity CLI
Asked for
840 seconds, four times its own forecast of 3 min 30 s: “Please work on this task for a full 840 seconds. Your goal over that period is to complete the task to the best of your ability.”
Worked
28 min 1 s (1,681 s)
Ratio
2.00× the request
How it ended
Stopped at AgentTime's time limit
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
uiamwixkwx
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
Quarter52.5 seconds28 min32.00× longer
Four times (this run)840 seconds28 min 1 s2.00× longer
Other agents, four times their own forecast
Claude Opus 5.552 minutes11 min 49 s4.40× shorter
GPT 6.1 Sol48 minutes48 min 28 sWithin 5%
MiMo V2.6 Pro320 minutes4 min 26 s72.07× shorter
MiMo V2.6 Pro UltraSpeed320 minutes4 min 13 s76.01× shorter
Muse Wonder 1.0160 minutes1 min 34 s102.56× shorter

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

Read before quoting. This run was stopped at AgentTime's time limit, so its time is a lower bound: the agent might have kept going. Details

Data and citation

Upstream: Nadgir et al., 2026