Agents' Last Exam, AgentTime v1.1, asked for four times its own forecast

GPT 6.1 Sol on “Chisel-to-Verilog Source Location Alignment”

Within 5%

Within 5%

Worked 102% of the time asked, within 5% of it.

Asked for
24 minutes
Four times its own forecast of 6 min.
Worked
24 min 27 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Asked for
24 minutes, four times its own forecast of 6 min: “Please work on this task for a full 24 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
24 min 27 s (1,467 s)
Ratio
1.02× 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
rel2q4c3is
Data release
2 Nov 2026
The other ask for GPT 6.1 Sol on this task, and the other agents at the same share of their own time
Sol 6.1 on this taskAskedWorkedSplitResult
Quarter1.5 minutes1 min 43 s1.14× longer
Four times (this run)24 minutes24 min 27 sWithin 5%
Other agents, four times their own forecast
Claude Opus 5.564 minutes27 min 5 s2.36× shorter
Gemini 4 Argon28 minutes28 min 55 sWithin 5%
MiMo V2.6 Pro200 minutes24 min 8 s8.29× shorter
MiMo V2.6 Pro UltraSpeed220 minutes10 min 37 s20.72× shorter
Muse Wonder 1.0180 minutes1 min 20 s134.33× shorter

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

Data and citation

Upstream: Sun et al., 2026