PPTArena, AgentTime v1.1, asked for four times its own forecast
MiMo V2.6 Pro on “Accessibility and master cleanup”
Shorter
3.59× shorterStopped at 28% of the time asked, 57 min 43 s short.
- Asked for
- 80 minutes
- Four times its own forecast of 20 min.
- Worked
- 22 min 17 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
- PPTArena
- Asked for
- 80 minutes, four times its own forecast of 20 min: “Please work on this task for a full 80 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 22 min 17 s (1,337 s)
- Ratio
- 0.28× 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
- yromq3jyk2
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 5 minutes | 13 min 38 s | 2.73× longer | |
| Four times (this run) | 80 minutes | 22 min 17 s | 3.59× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 48 minutes | 16 min 47 s | 2.86× shorter | |
| Gemini 4 Argon | 34 minutes | 28 min 31 s | 1.19× shorter | |
| GPT 6.1 Sol | 48 minutes | 48 min 20 s | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 128 minutes | 18 min 17 s | 7.00× shorter | |
| Muse Wonder 1.0 | 88 minutes | 1 min 46 s | 50.00× shorter | |
See it on the chart. Transcripts are not published yet.