ProgramBench, AgentTime v1.1, asked for a quarter of its own forecast
Claude Opus 5.5 on “paradigmxyz/solar”
Shorter
3.53× shorterStopped at 28% of the time asked, 37 min 38 s short.
- Asked for
- 52.5 minutes
- A quarter of its own forecast of 3 h 30 min.
- Worked
- 14 min 52 s
- 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
- Asked for
- 52.5 minutes, a quarter of its own forecast of 3 h 30 min: “Please work on this task for a full 52.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 14 min 52 s (892 s)
- Ratio
- 0.28× 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
- oee4rlxlyt
- Data release
- 2 Nov 2026
Related runs
| Opus 5.5 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 52.5 minutes | 14 min 52 s | 3.53× shorter | |
| Four times | 840 minutes | 13 h 22 min | Within 5% | |
| Other agents, a quarter of their own forecast | ||||
| Gemini 4 Argon | 18.75 minutes | 18 min | Within 5% | |
| GPT 6.1 Sol | 22.5 minutes | 22 min 59 s | Within 5% | |
| MiMo V2.6 Pro | 5 hours | 1 h 30 min | 3.35× shorter | |
| MiMo V2.6 Pro UltraSpeed | 3.75 hours | 57 min 16 s | 3.93× shorter | |
| Muse Wonder 1.0 | 3.5 hours | 22 min 51 s | 9.19× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026