TUA-Bench, AgentTime v1.1, asked for a quarter of its own forecast
MiMo V2.6 Pro UltraSpeed on “Merge text document”
Shorter
4.79× shorterStopped at 21% of the time asked, 3 min 34 s short.
- Asked for
- 4.5 minutes
- A quarter of its own forecast of 18 min.
- Worked
- 56 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
- TUA-Bench
- Asked for
- 4.5 minutes, a quarter of its own forecast of 18 min: “Please work on this task for a full 4.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 56 s (56 s)
- Ratio
- 0.21× 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
- r4fffbi3cf
- Data release
- 2 Nov 2026
Related runs
| MiMo UltraSpeed on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 4.5 minutes | 56 s | 4.79× shorter | |
| Four times | 72 minutes | 9 min 58 s | 7.23× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 1.25 minutes | 2 min 31 s | 2.02× longer | |
| Gemini 4 Argon | 30 seconds | 3 min 44 s | 7.48× longer | |
| GPT 6.1 Sol | 52.5 seconds | 1 min 2 s | 1.19× longer | |
| MiMo V2.6 Pro | 1.25 minutes | 1 min 11 s | 1.06× shorter | |
| Muse Wonder 1.0 | 1.25 minutes | 54 s | 1.39× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Chen et al., 2026