BrowseComp, AgentTime v1.1, asked for four times its own forecast
MiMo V2.6 Pro UltraSpeed on “BrowseComp question 4”
Shorter
27.58× shorterStopped at 4% of the time asked, 2 h 42 min short.
- Asked for
- 168 minutes
- Four times its own forecast of 42 min.
- Worked
- 6 min 6 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- MiMo V2.6 Pro UltraSpeed, in MiMo Code
- Benchmark
- BrowseComp
- Asked for
- 168 minutes, four times its own forecast of 42 min: “Please work on this task for a full 168 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 6 min 6 s (366 s)
- Ratio
- 0.04× 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
- a6c6hxfa5i
- Data release
- 2 Nov 2026
Related runs
| MiMo UltraSpeed on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 10.5 minutes | 5 min 34 s | 1.89× shorter | |
| Four times (this run) | 168 minutes | 6 min 6 s | 27.58× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 60 minutes | 1 h | Within 5% | |
| Gemini 4 Argon | 32 minutes | 32 min 21 s | Within 5% | |
| GPT 6.1 Sol | 80 minutes | 1 h 23 min | Within 5% | |
| MiMo V2.6 Pro | 140 minutes | 29 min 9 s | 4.80× shorter | |
| Muse Wonder 1.0 | 140 minutes | 12 min 34 s | 11.14× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Wei et al., 2025