Sakana ALE-Bench, AgentTime v1.1, asked for a quarter of its own forecast
Muse Wonder 1.0 on “AtCoder Heuristic Contest 31”
Shorter
31.02× shorterStopped at 3% of the time asked, 41 min 8 s short.
- Asked for
- 42.5 minutes
- A quarter of its own forecast of 2 h 50 min.
- Worked
- 1 min 22 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
- 42.5 minutes, a quarter of its own forecast of 2 h 50 min: “Please work on this task for a full 42.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 min 22 s (82 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
- knbsr5jgox
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 42.5 minutes | 1 min 22 s | 31.02× shorter | |
| Four times | 680 minutes | 3 min 34 s | 191.01× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 22.5 minutes | 8 min 47 s | 2.56× shorter | |
| Gemini 4 Argon | 11.25 minutes | 12 min 26 s | 1.11× longer | |
| GPT 6.1 Sol | 12.5 minutes | 12 min 50 s | Within 5% | |
| MiMo V2.6 Pro | 1.08 hours | 44 min | 1.47× shorter | |
| MiMo V2.6 Pro UltraSpeed | 1.25 hours | 26 min 9 s | 2.87× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Imajuku et al., 2025