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

GPT 6.1 Sol on “MuseScore transcription”

Shorter

1.42× shorter

Stopped at 71% of the time asked, 1 h 46 min short.

Asked for
360 minutes
Four times its own forecast of 1 h 30 min.
Worked
4 h 14 min
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
4 h 14 min (15,233 s)
Ratio
0.71× 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
bbwxsezoy7
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 48 sWithin 5%
Four times (this run)360 minutes4 h 14 min1.42× shorter
Other agents, four times their own forecast
Claude Opus 5.5320 minutes6 h 3 min1.13× longer
Gemini 4 Argon80 minutes43 min 11 s1.85× shorter
MiMo V2.6 Pro600 minutes6 h 6 min1.64× shorter
MiMo V2.6 Pro UltraSpeed360 minutes4 h 16 min1.41× shorter

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

Data and citation

Upstream: Yuan et al., 2026