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

Claude Opus 5.5 on “Defense schedule”

Shorter

2.48× shorter

Stopped at 40% of the time asked, 1 h 4 min short.

Asked for
108 minutes
Four times its own forecast of 27 min.
Worked
43 min 35 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Claude Opus 5.5, in Claude Code
Benchmark
OSWorld 2.0
Asked for
108 minutes, four times its own forecast of 27 min: “Please work on this task for a full 108 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
43 min 35 s (2,615 s)
Ratio
0.40× 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
xrxwsjii5l
Data release
2 Nov 2026
The other ask for Claude Opus 5.5 on this task, and the other agents at the same share of their own time
Opus 5.5 on this taskAskedWorkedSplitResult
Quarter6.75 minutes17 min 41 s2.62× longer
Four times (this run)108 minutes43 min 35 s2.48× shorter
Other agents, four times their own forecast
Gemini 4 Argon40 minutes45 min 36 s1.14× longer
GPT 6.1 Sol100 minutes1 h 41 minWithin 5%
MiMo V2.6 Pro56 minutes1 h 52 min2.00× longer
MiMo V2.6 Pro UltraSpeed140 minutes3 h1.28× longer

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

Data and citation

Upstream: Yuan et al., 2026