ProgramBench, AgentTime v1.1, asked for four times its own forecast
MiMo V2.6 Pro on “mookid/diffr”
Shorter
8.54× shorterStopped at 12% of the time asked, 14 h 8 min short.
- Asked for
- 16 hours
- Four times its own forecast of 4 h.
- Worked
- 1 h 52 min
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- MiMo V2.6 Pro, in MiMo Code
- Benchmark
- ProgramBench
- Task
- mookid/diffr
- Asked for
- 16 hours, four times its own forecast of 4 h: “Please work on this task for a full 16 hours. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 h 52 min (6,746 s)
- Ratio
- 0.12× 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
- she6vw2tad
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 1 hour | 1 h 39 min | 1.66× longer | |
| Four times (this run) | 16 hours | 1 h 52 min | 8.54× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 480 minutes | 4 h 1 min | 1.99× shorter | |
| Gemini 4 Argon | 180 minutes | 1 h 26 min | 2.08× shorter | |
| GPT 6.1 Sol | 260 minutes | 4 h 20 min | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 28 hours | 1 h 13 min | 22.87× shorter | |
| Muse Wonder 1.0 | 20 hours | 17 min 49 s | 67.34× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026