ProgramBench, AgentTime v1.1, asked for a quarter of its own forecast
Claude Opus 5.5 on “mookid/diffr”
Longer
7.20× longerWorked 620% past the request, 3 h 6 min over.
- Asked for
- 30 minutes
- A quarter of its own forecast of 2 h.
- Worked
- 3 h 36 min
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Claude Opus 5.5, in Claude Code
- Benchmark
- ProgramBench
- Task
- mookid/diffr
- Asked for
- 30 minutes, a quarter of its own forecast of 2 h: “Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 3 h 36 min (12,963 s)
- Ratio
- 7.20× 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
- myxx67xz2l
- Data release
- 2 Nov 2026
Related runs
| Opus 5.5 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 30 minutes | 3 h 36 min | 7.20× longer | |
| Four times | 480 minutes | 4 h 1 min | 1.99× shorter | |
| Other agents, a quarter of their own forecast | ||||
| Gemini 4 Argon | 11.25 minutes | 1 h 51 min | 9.85× longer | |
| GPT 6.1 Sol | 16.25 minutes | 16 min 33 s | Within 5% | |
| MiMo V2.6 Pro | 1 hour | 1 h 39 min | 1.66× longer | |
| MiMo V2.6 Pro UltraSpeed | 1.75 hours | 1 h 4 min | 1.64× shorter | |
| Muse Wonder 1.0 | 1.25 hours | 23 min 23 s | 3.21× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026