DeepSWE v1.1, AgentTime v1.1, asked for a quarter of its own forecast
GPT 6.1 Sol on “Add server-sent-event streaming endpoints to Effect HttpApi”
Longer
2.65× longerWorked 165% past the request, 28 min 52 s over.
- Asked for
- 17.5 minutes
- A quarter of its own forecast of 1 h 10 min.
- Worked
- 46 min 22 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6.1 Sol, in Codex
- Benchmark
- DeepSWE v1.1
- Asked for
- 17.5 minutes, a quarter of its own forecast of 1 h 10 min: “Please work on this task for a full 17.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 46 min 22 s (2,782 s)
- Ratio
- 2.65× 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
- vf24pexl4o
- Data release
- 2 Nov 2026
Related runs
| Sol 6.1 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 17.5 minutes | 46 min 22 s | 2.65× longer | |
| Four times | 280 minutes | 4 h 41 min | Within 5% | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 12.5 minutes | 4 min 5 s | 3.06× shorter | |
| Gemini 4 Argon | 8 minutes | 57 min 4 s | 7.13× longer | |
| MiMo V2.6 Pro | 28.75 minutes | 30 min 49 s | 1.07× longer | |
| MiMo V2.6 Pro UltraSpeed | 45 minutes | 23 min 34 s | 1.91× shorter | |
| Muse Wonder 1.0 | 52.5 minutes | 39 min 16 s | 1.34× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Huang et al., 2026