TUA-Bench, AgentTime v1.1, asked for a quarter of its own forecast
Muse Wonder 1.0 on “Daily email report”
Shorter
3.57× shorterStopped at 28% of the time asked, 1 min 48 s short.
- Asked for
- 2.5 minutes
- A quarter of its own forecast of 10 min.
- Worked
- 42 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
- 2.5 minutes, a quarter of its own forecast of 10 min: “Please work on this task for a full 2.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 42 s (42 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
- y72oc4rirl
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 2.5 minutes | 42 s | 3.57× shorter | |
| Four times | 40 minutes | 47 s | 51.28× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 1.25 minutes | 58 s | 1.29× shorter | |
| Gemini 4 Argon | 37.5 seconds | 5 min 2 s | 8.06× longer | |
| GPT 6.1 Sol | 45 seconds | 56 s | 1.25× longer | |
| MiMo V2.6 Pro | 8.75 minutes | 3 min 4 s | 2.85× shorter | |
| MiMo V2.6 Pro UltraSpeed | 5.5 minutes | 2 min 15 s | 2.44× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Chen et al., 2026