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

GPT 6.1 Sol on “Calc macros, contracts, and emails”

Shorter

3.97× shorter

Stopped at 25% of the time asked, 3 h short.

Asked for
240 minutes
Four times its own forecast of 1 h.
Worked
1 h
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
240 minutes, four times its own forecast of 1 h: “Please work on this task for a full 240 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
1 h (3,625 s)
Ratio
0.25× 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
3b6izevrfb
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
Quarter15 minutes15 min 35 sWithin 5%
Four times (this run)240 minutes1 h3.97× shorter
Other agents, four times their own forecast
Claude Opus 5.5180 minutes2 h 16 min1.32× shorter
Gemini 4 Argon80 minutes34 min 47 s2.30× shorter
MiMo V2.6 Pro220 minutes6 h 41 min1.82× longer
MiMo V2.6 Pro UltraSpeed360 minutes6 h 31 min1.09× longer

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

Data and citation

Upstream: Yuan et al., 2026