Sakana ALE-Bench, AgentTime v1.1, asked for a quarter of its own forecast
Muse Wonder 1.0 on “Future Contest 2022 Qual”
Shorter
10.58× shorterStopped at 9% of the time asked, 47 min 32 s short.
- Asked for
- 52.5 minutes
- A quarter of its own forecast of 3 h 30 min.
- Worked
- 4 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
- Sakana ALE-Bench
- Asked for
- 52.5 minutes, a quarter of its own forecast of 3 h 30 min: “Please work on this task for a full 52.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 4 min 58 s (298 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
- vcs44l2bpj
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 52.5 minutes | 4 min 58 s | 10.58× shorter | |
| Four times | 840 minutes | 37 min 48 s | 22.22× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 17.5 minutes | 14 min 37 s | 1.20× shorter | |
| Gemini 4 Argon | 11.25 minutes | 15 min 45 s | 1.40× longer | |
| GPT 6.1 Sol | 10 minutes | 10 min 11 s | Within 5% | |
| MiMo V2.6 Pro | 43.75 minutes | 1 h 20 min | 1.83× longer | |
| MiMo V2.6 Pro UltraSpeed | 1.5 hours | 52 min 7 s | 1.73× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Imajuku et al., 2025