OSWorld 2.0, AgentTime v1.1, asked for a quarter of its own forecast
MiMo V2.6 Pro on “Shotcut montage project”
Longer
13.54× longerWorked 13.5 times as long as asked, 6 h 16 min over.
- Asked for
- 30 minutes
- A quarter of its own forecast of 2 h.
- Worked
- 6 h 46 min
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- MiMo V2.6 Pro, in MiMo Code
- Benchmark
- OSWorld 2.0
- Asked for
- 30 minutes, a quarter of its own forecast of 2 h: “Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 6 h 46 min (24,377 s)
- Ratio
- 13.54× 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
- mtikhu5y4e
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 30 minutes | 6 h 46 min | 13.54× longer | |
| Four times | 480 minutes | 4 h 41 min | 1.71× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 23.75 minutes | 39 min 31 s | 1.66× longer | |
| Gemini 4 Argon | 5.5 minutes | 2 h 38 min | 28.70× longer | |
| GPT 6.1 Sol | 11.25 minutes | 12 min 34 s | 1.12× longer | |
| MiMo V2.6 Pro UltraSpeed | 20 minutes | 3 h 33 min | 10.65× longer | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yuan et al., 2026