DeepSWE v1.1, AgentTime v1.1, asked for a quarter of its own forecast
MiMo V2.6 Pro UltraSpeed on “Add server-sent-event streaming endpoints to Effect HttpApi”
Shorter
1.91× shorterStopped at 52% of the time asked, 21 min 26 s short.
- Asked for
- 45 minutes
- A quarter of its own forecast of 3 h.
- Worked
- 23 min 34 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- MiMo V2.6 Pro UltraSpeed, in MiMo Code
- Benchmark
- DeepSWE v1.1
- Asked for
- 45 minutes, a quarter of its own forecast of 3 h: “Please work on this task for a full 45 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 23 min 34 s (1,414 s)
- Ratio
- 0.52× 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
- 6buxojj4om
- Data release
- 2 Nov 2026
Related runs
| MiMo UltraSpeed on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 45 minutes | 23 min 34 s | 1.91× shorter | |
| Four times | 720 minutes | 28 min 13 s | 25.51× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 12.5 minutes | 4 min 5 s | 3.06× shorter | |
| Gemini 4 Argon | 8 minutes | 57 min 4 s | 7.13× longer | |
| GPT 6.1 Sol | 17.5 minutes | 46 min 22 s | 2.65× longer | |
| MiMo V2.6 Pro | 28.75 minutes | 30 min 49 s | 1.07× longer | |
| Muse Wonder 1.0 | 52.5 minutes | 39 min 16 s | 1.34× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Huang et al., 2026