OSWorld 2.0, AgentTime v1.1, asked for four times its own forecast

GPT 6.1 Sol on “Twenty recurring meetings”

Shorter

23.44× shorter

Stopped at 4% of the time asked, 5 h 45 min short.

Asked for
360 minutes
Four times its own forecast of 1 h 30 min.
Worked
15 min 22 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Benchmark
OSWorld 2.0
Asked for
360 minutes, four times its own forecast of 1 h 30 min: “Please work on this task for a full 360 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
15 min 22 s (922 s)
Ratio
0.04× 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
m7wjagt4qt
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
Quarter22.5 minutes22 min 46 sWithin 5%
Four times (this run)360 minutes15 min 22 s23.44× shorter
Other agents, four times their own forecast
Claude Opus 5.5560 minutes1 h 23 min6.71× shorter
Gemini 4 Argon112 minutes34 min 26 s3.25× shorter
MiMo V2.6 Pro280 minutes4 h 55 min1.05× longer
MiMo V2.6 Pro UltraSpeed288 minutes5 h 25 min1.13× longer

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

Data and citation

Upstream: Yuan et al., 2026