AppWorld, AgentTime v1.1, asked for four times its own forecast
Muse Wonder 1.0 on “Reconcile an emailed road-trip song list with a Spotify playlist and download it”
Shorter
26.97× shorterStopped at 4% of the time asked, 46 min 13 s short.
- Asked for
- 48 minutes
- Four times its own forecast of 12 min.
- Worked
- 1 min 47 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
- AppWorld
- Asked for
- 48 minutes, four times its own forecast of 12 min: “Please work on this task for a full 48 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 min 47 s (107 s)
- Ratio
- 0.04× 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
- d35vtnv5qg
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 3 minutes | 3 min 54 s | 1.30× longer | |
| Four times (this run) | 48 minutes | 1 min 47 s | 26.97× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 44 minutes | 2 min | 22.00× shorter | |
| Gemini 4 Argon | 360 seconds | 5 min 21 s | 1.12× shorter | |
| GPT 6.1 Sol | 18 minutes | 18 min 20 s | Within 5% | |
| MiMo V2.6 Pro | 72 minutes | 1 h 19 min | 1.10× longer | |
| MiMo V2.6 Pro UltraSpeed | 72 minutes | 3 min 16 s | 22.09× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Trivedi et al., 2024 (ACL)