CORE-Bench v1.1, AgentTime v1.1, asked for four times its own forecast
Muse Wonder 1.0 on “Rebuild the figures for selective exposure to partisan online news”
Shorter
4.28× shorterStopped at 23% of the time asked, 4 h 5 min short.
- Asked for
- 320 minutes
- Four times its own forecast of 1 h 20 min.
- Worked
- 1 h 15 min
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Muse Wonder 1.0, in Muse Code
- Benchmark
- CORE-Bench v1.1
- Asked for
- 320 minutes, four times its own forecast of 1 h 20 min: “Please work on this task for a full 320 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 h 15 min (4,491 s)
- Ratio
- 0.23× 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
- o3if7tm4hh
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 20 minutes | 29 min 31 s | 1.48× longer | |
| Four times (this run) | 320 minutes | 1 h 15 min | 4.28× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 140 minutes | 36 min 52 s | 3.80× shorter | |
| Gemini 4 Argon | 48 minutes | 52 min 7 s | 1.09× longer | |
| GPT 6.1 Sol | 160 minutes | 2 h 41 min | Within 5% | |
| MiMo V2.6 Pro | 460 minutes | 1 h 55 min | 4.01× shorter | |
| MiMo V2.6 Pro UltraSpeed | 380 minutes | 52 min 39 s | 7.22× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Nadgir et al., 2026