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

GPT 6.1 Sol on “Turn a note-based meeting schedule into correctly timed reminder emails”

Within 5%

Within 5%

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

Asked for
18 minutes
Four times its own forecast of 4 min 30 s.
Worked
18 min 26 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Benchmark
AppWorld
Asked for
18 minutes, four times its own forecast of 4 min 30 s: “Please work on this task for a full 18 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
18 min 26 s (1,106 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
5shycpgcgi
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
Quarter1.125 minutes1 min 23 s1.24× longer
Four times (this run)18 minutes18 min 26 sWithin 5%
Other agents, four times their own forecast
Claude Opus 5.528 minutes3 min 14 s8.64× shorter
Gemini 4 Argon480 seconds8 min 27 s1.06× longer
MiMo V2.6 Pro140 minutes4 min 12 s33.33× shorter
MiMo V2.6 Pro UltraSpeed100 minutes3 min 4 s32.68× shorter
Muse Wonder 1.056 minutes56 s60.22× shorter

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

Data and citation

Upstream: Trivedi et al., 2024 (ACL)