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

GPT 6.1 Sol on “Buy a requested item, forward its receipt, and recover the cost through Venmo”

Within 5%

Within 5%

Worked 103% of the time asked, within 5% of it.

Asked for
20 minutes
Four times its own forecast of 5 min.
Worked
20 min 37 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
20 minutes, four times its own forecast of 5 min: “Please work on this task for a full 20 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
20 min 37 s (1,237 s)
Ratio
1.03× 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
fmparjsqpw
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
Quarter1.25 minutes1 min 26 s1.15× longer
Four times (this run)20 minutes20 min 37 sWithin 5%
Other agents, four times their own forecast
Claude Opus 5.528 minutes6 min 5 s4.61× shorter
Gemini 4 Argon600 seconds20 min2.00× longer
MiMo V2.6 Pro64 minutes34 min 11 s1.87× shorter
MiMo V2.6 Pro UltraSpeed112 minutes18 min 36 s6.02× shorter
Muse Wonder 1.064 minutes11 s355.56× shorter

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

Data and citation

Upstream: Trivedi et al., 2024 (ACL)