OSWorld 2.0, AgentTime v1.1, asked for a quarter of its own forecast
MiMo V2.6 Pro on “Defense schedule”
Longer
30.58× longerWorked 30.6 times as long as asked, 1 h 44 min over.
- Asked for
- 3.5 minutes
- A quarter of its own forecast of 14 min.
- Worked
- 1 h 47 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
- Task
- Defense schedule
- Asked for
- 3.5 minutes, a quarter of its own forecast of 14 min: “Please work on this task for a full 3.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 h 47 min (6,421 s)
- Ratio
- 30.58× 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
- ga5oeujhwv
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 3.5 minutes | 1 h 47 min | 30.58× longer | |
| Four times | 56 minutes | 1 h 52 min | 2.00× longer | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 6.75 minutes | 17 min 41 s | 2.62× longer | |
| Gemini 4 Argon | 2.5 minutes | 15 min 53 s | 6.36× longer | |
| GPT 6.1 Sol | 6.25 minutes | 6 min 23 s | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 8.75 minutes | 3 h 20 min | 22.90× longer | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yuan et al., 2026