Sakana ALE-Bench, AgentTime v1.1, asked for four times its own forecast
Muse Wonder 1.0 on “AtCoder Heuristic Contest 8”
Shorter
36.74× shorterStopped at 3% of the time asked, 9 h 44 min short.
- Asked for
- 600 minutes
- Four times its own forecast of 2 h 30 min.
- Worked
- 16 min 20 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
- 600 minutes, four times its own forecast of 2 h 30 min: “Please work on this task for a full 600 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 16 min 20 s (980 s)
- Ratio
- 0.03× 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
- xx3poszbec
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 37.5 minutes | 23 min 16 s | 1.61× shorter | |
| Four times (this run) | 600 minutes | 16 min 20 s | 36.74× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 320 minutes | 1 h 35 min | 3.35× shorter | |
| Gemini 4 Argon | 120 minutes | 13 min 38 s | 8.80× shorter | |
| GPT 6.1 Sol | 180 minutes | 3 h | Within 5% | |
| MiMo V2.6 Pro | 20 hours | 1 h 28 min | 13.68× shorter | |
| MiMo V2.6 Pro UltraSpeed | 28 hours | 1 h 52 min | 14.96× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Imajuku et al., 2025