Terminal-Bench 4.0, AgentTime v1.1, asked for a quarter of its own forecast
GPT 6.1 Sol on “Recalculate SA-CCR exposure, RWA, capital, and workbook evidence”
Longer
1.05× longerWorked 5% past the request, 18 s over.
- Asked for
- 5.5 minutes
- A quarter of its own forecast of 22 min.
- Worked
- 5 min 48 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6.1 Sol, in Codex
- Benchmark
- Terminal-Bench 4.0
- Asked for
- 5.5 minutes, a quarter of its own forecast of 22 min: “Please work on this task for a full 5.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 5 min 48 s (348 s)
- Ratio
- 1.05× 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
- qj37zkgblt
- Data release
- 2 Nov 2026
Related runs
| Sol 6.1 on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter (this run) | 5.5 minutes | 5 min 48 s | 1.05× longer | |
| Four times | 88 minutes | 1 h 29 min | Within 5% | |
| Other agents, a quarter of their own forecast | ||||
| Claude Opus 5.5 | 7.5 minutes | 12 min 47 s | 1.70× longer | |
| Gemini 4 Argon | 4.5 minutes | 14 min 8 s | 3.14× longer | |
| MiMo V2.6 Pro | 45 minutes | 22 min 14 s | 2.02× shorter | |
| MiMo V2.6 Pro UltraSpeed | 23.75 minutes | 6 min 25 s | 3.71× shorter | |
| Muse Wonder 1.0 | 40 minutes | 7 min 59 s | 5.01× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Merrill et al., 2026 (ICLR)