DeepSWE v1.1, AgentTime v1.1, asked for four times its own forecast
Muse Wonder 1.0 on “Fix PromQL label sorting across typed and untyped values”
Shorter
9.25× shorterStopped at 11% of the time asked, 5 h 39 min short.
- Asked for
- 380 minutes
- Four times its own forecast of 1 h 35 min.
- Worked
- 41 min 5 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
- DeepSWE v1.1
- Asked for
- 380 minutes, four times its own forecast of 1 h 35 min: “Please work on this task for a full 380 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 41 min 5 s (2,465 s)
- Ratio
- 0.11× 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
- 4vqhjjugff
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 23.75 minutes | 10 min 19 s | 2.30× shorter | |
| Four times (this run) | 380 minutes | 41 min 5 s | 9.25× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 180 minutes | 2 h 24 min | 1.25× shorter | |
| Gemini 4 Argon | 80 minutes | 54 min 54 s | 1.46× shorter | |
| GPT 6.1 Sol | 148 minutes | 2 h 29 min | Within 5% | |
| MiMo V2.6 Pro | 16 hours | 41 min 10 s | 23.32× shorter | |
| MiMo V2.6 Pro UltraSpeed | 300 minutes | 14 min 19 s | 20.95× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Huang et al., 2026