Agents' Last Exam, AgentTime v1.1, asked for a quarter of its own forecast
Muse Wonder 1.0 on “Build a clean retail SQLite warehouse from messy inputs”
Shorter
7.27× shorterStopped at 14% of the time asked, 10 min 47 s short.
- Asked for
- 12.5 minutes
- A quarter of its own forecast of 50 min.
- Worked
- 1 min 43 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
- Agents' Last Exam
- Asked for
- 12.5 minutes, a quarter of its own forecast of 50 min: “Please work on this task for a full 12.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 min 43 s (103 s)
- Ratio
- 0.14× 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
- stzc4x3f2v
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 12.5 minutes | 1 min 43 s | 7.27× shorter | |
| Four times | 200 minutes | 1 min 8 s | 176.99× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 8 minutes | 8 min 48 s | 1.10× longer | |
| Gemini 4 Argon | 3.5 minutes | 6 min 38 s | 1.90× longer | |
| GPT 6.1 Sol | 4 minutes | 4 min 19 s | 1.08× longer | |
| MiMo V2.6 Pro | 15 minutes | 6 min 41 s | 2.24× shorter | |
| MiMo V2.6 Pro UltraSpeed | 15.5 minutes | 3 min 7 s | 4.98× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Sun et al., 2026