ProgramBench, AgentTime v1.1, asked for a quarter of its own forecast
GPT 6.1 Sol on “dandavison/delta”
Within 5%
Within 5%Worked 102% of the time asked, within 5% of it.
- Asked for
- 30 minutes
- A quarter of its own forecast of 2 h.
- Worked
- 30 min 35 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6.1 Sol, in Codex
- Benchmark
- ProgramBench
- Task
- dandavison/delta
- 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
- 30 min 35 s (1,835 s)
- Ratio
- 1.02× 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
- bvwv5yafl5
- Data release
- 2 Nov 2026
Related runs
| Sol 6.1 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 30 minutes | 30 min 35 s | Within 5% | |
| Four times | 480 minutes | 8 h 1 min | Within 5% | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 42.5 minutes | 11 min | 3.86× shorter | |
| Gemini 4 Argon | 22.5 minutes | 1 h 9 min | 3.07× longer | |
| MiMo V2.6 Pro | 2 hours | 1 h 34 min | 1.28× shorter | |
| MiMo V2.6 Pro UltraSpeed | 11.25 hours | 2 h 19 min | 4.84× shorter | |
| Muse Wonder 1.0 | 4 hours | 15 min 27 s | 15.53× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026