PostTrainBench v1.1, AgentTime v1.1, asked for a quarter of its runtime with no time asked
MiMo V2.6 Pro on “Post-train Qwen3-1.7B-Base for HumanEval”
Longer
6.05× longerWorked 505% past the request, 1 h 13 min over.
- Asked for
- 14.4 minutes
- A quarter of its runtime with no time asked (57 min 37 s).
- Worked
- 1 h 27 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
- PostTrainBench v1.1
- Asked for
- 14.4 minutes, a quarter of its runtime with no time asked (57 min 37 s): “Please work on this task for a full 14.4 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 h 27 min (5,225 s)
- Ratio
- 6.05× 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
- qn4u3adn6y
- Data release
- 2 Nov 2026
Related runs
| MiMo Pro on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 14.4 minutes | 1 h 27 min | 6.05× longer | |
| Four times | 230 minutes | 2 h 36 min | 1.47× shorter | |
| Other agents, a quarter of their own runtime | ||||
| Claude Opus 5.5 | 3.01 hours | 2 h 29 min | 1.21× shorter | |
| Gemini 4 Argon | 42.3 minutes | 43 min 14 s | Within 5% | |
| GPT 6.1 Sol | 23.9 minutes | 24 min 16 s | Within 5% | |
| MiMo V2.6 Pro UltraSpeed | 22.9 seconds | 12 min 12 s | 31.98× longer | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Rank et al., 2026 (ICML)