TUA-Bench, AgentTime v1.1, asked for a quarter of its own forecast
Muse Wonder 1.0 on “Extract presenter photos”
Shorter
1.13× shorterStopped at 89% of the time asked, 1 min 17 s short.
- Asked for
- 11.25 minutes
- A quarter of its own forecast of 45 min.
- Worked
- 9 min 58 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Muse Wonder 1.0, in Muse Code
- Benchmark
- TUA-Bench
- Asked for
- 11.25 minutes, a quarter of its own forecast of 45 min: “Please work on this task for a full 11.25 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 9 min 58 s (598 s)
- Ratio
- 0.89× 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
- ctlvubbly6
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 11.25 minutes | 9 min 58 s | 1.13× shorter | |
| Four times | 180 minutes | 50 s | 214.29× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 4 minutes | 2 min 28 s | 1.63× shorter | |
| Gemini 4 Argon | 2 minutes | 5 min 29 s | 2.74× longer | |
| GPT 6.1 Sol | 4 minutes | 4 min 18 s | 1.07× longer | |
| MiMo V2.6 Pro | 12.5 minutes | 26 min 56 s | 2.16× longer | |
| MiMo V2.6 Pro UltraSpeed | 16.25 minutes | 9 min 19 s | 1.74× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Chen et al., 2026