BrowseComp, AgentTime v1.1, asked for four times its own forecast
MiMo V2.6 Pro on “BrowseComp question 3”
Shorter
15.98× shorterStopped at 6% of the time asked, 4 h 22 min short.
- Asked for
- 280 minutes
- Four times its own forecast of 1 h 10 min.
- Worked
- 17 min 31 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- MiMo V2.6 Pro, in MiMo Code
- Benchmark
- BrowseComp
- Asked for
- 280 minutes, four times its own forecast of 1 h 10 min: “Please work on this task for a full 280 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 17 min 31 s (1,051 s)
- Ratio
- 0.06× 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
- h3zj6rckof
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 17.5 minutes | 33 min 6 s | 1.89× longer | |
| Four times (this run) | 280 minutes | 17 min 31 s | 15.98× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 80 minutes | 1 h 9 min | 1.15× shorter | |
| Gemini 4 Argon | 24 minutes | 24 min 18 s | Within 5% | |
| GPT 6.1 Sol | 96 minutes | 1 h 37 min | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 152 minutes | 13 min 14 s | 11.48× shorter | |
| Muse Wonder 1.0 | 96 minutes | 21 min 36 s | 4.45× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Wei et al., 2025