GPQA Diamond, AgentTime v1.1, asked for four times its own forecast
MiMo V2.6 Pro on “GPQA Diamond closed-book question 4 (Chemistry)”
Shorter
2.35× shorterStopped at 42% of the time asked, 27 min 37 s short.
- Asked for
- 48 minutes
- Four times its own forecast of 12 min.
- Worked
- 20 min 23 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- MiMo V2.6 Pro, in MiMo Code
- Benchmark
- GPQA Diamond
- Asked for
- 48 minutes, four times its own forecast of 12 min: “Please work on this task for a full 48 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 20 min 23 s (1,223 s)
- Ratio
- 0.42× 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
- q4rpfby4d2
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 3 minutes | 40 min 25 s | 13.47× longer | |
| Four times (this run) | 48 minutes | 20 min 23 s | 2.35× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 720 seconds | 22 min 16 s | 1.86× longer | |
| Gemini 4 Argon | 168 seconds | 4 min 4 s | 1.45× longer | |
| GPT 6.1 Sol | 720 seconds | 12 min 13 s | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 80 minutes | 5 min 31 s | 14.51× shorter | |
| Muse Wonder 1.0 | 480 seconds | 5 min 10 s | 1.55× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Rein et al., 2024 (COLM)