ProgramBench, AgentTime v1.1, asked for four times its own forecast
Muse Wonder 1.0 on “stranger6667/jsonschema”
Shorter
159.90× shorterStopped at 1% of the time asked, 43 h 43 min short.
- Asked for
- 44 hours
- Four times its own forecast of 11 h.
- Worked
- 16 min 31 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
- ProgramBench
- Asked for
- 44 hours, four times its own forecast of 11 h: “Please work on this task for a full 44 hours. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 16 min 31 s (991 s)
- Ratio
- 0.01× 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
- 4serbx25id
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 2.75 hours | 16 min 47 s | 9.83× shorter | |
| Four times (this run) | 44 hours | 16 min 31 s | 159.90× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 680 minutes | 10 h 56 min | Within 5% | |
| Gemini 4 Argon | 360 minutes | 1 h 48 min | 3.34× shorter | |
| GPT 6.1 Sol | 240 minutes | 4 h | Within 5% | |
| MiMo V2.6 Pro | 52 hours | 1 h 56 min | 26.83× shorter | |
| MiMo V2.6 Pro UltraSpeed | 28 hours | 1 h 15 min | 22.35× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026