Humanity's Last Exam, AgentTime v1.1, asked for four times its own forecast
MiMo V2.6 Pro on “Humanity's Last Exam item 25”
Shorter
5.90× shorterStopped at 17% of the time asked, 9 min 58 s short.
- Asked for
- 720 seconds
- Four times its own forecast of 3 min.
- Worked
- 2 min 2 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
- Humanity's Last Exam
- Asked for
- 720 seconds, four times its own forecast of 3 min: “Please work on this task for a full 720 seconds. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 2 min 2 s (122 s)
- Ratio
- 0.17× 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
- evjysw6my6
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 45 seconds | 47 s | Within 5% | |
| Four times (this run) | 720 seconds | 2 min 2 s | 5.90× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 360 seconds | 8 min 51 s | 1.47× longer | |
| Gemini 4 Argon | 120 seconds | 2 min 35 s | 1.29× longer | |
| GPT 6.1 Sol | 360 seconds | 6 min 16 s | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 60 minutes | 35 s | 103.15× shorter | |
| Muse Wonder 1.0 | 480 seconds | 1 min 29 s | 5.41× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Phan et al., 2025