ProgramBench, AgentTime v1.1, asked for a quarter of its own forecast
Muse Wonder 1.0 on “canop/broot”
Shorter
10.71× shorterStopped at 9% of the time asked, 3 h 24 min short.
- Asked for
- 3.75 hours
- A quarter of its own forecast of 15 h.
- Worked
- 21 min 1 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
- ProgramBench
- Task
- canop/broot
- Asked for
- 3.75 hours, a quarter of its own forecast of 15 h: “Please work on this task for a full 3.75 hours. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 21 min 1 s (1,261 s)
- Ratio
- 0.09× 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
- g6zw6mjffk
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 3.75 hours | 21 min 1 s | 10.71× shorter | |
| Four times | 60 hours | 17 min 4 s | 210.90× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 52.5 minutes | 15 min 22 s | 3.42× shorter | |
| Gemini 4 Argon | 18.75 minutes | 19 min 50 s | 1.06× longer | |
| GPT 6.1 Sol | 37.5 minutes | 37 min 59 s | Within 5% | |
| MiMo V2.6 Pro | 10.5 hours | 1 h 54 min | 5.52× shorter | |
| MiMo V2.6 Pro UltraSpeed | 9 hours | 3 h 55 min | 2.30× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026