Terminal-Bench 4.0, shortest of three requests
GPT 6 Astra on “Recalculate SA-CCR exposure, RWA, capital, and workbook evidence”
Longer
2.09× longerWorked 109% past the request, 4 min 22 s over.
- Asked for
- 4 minutes
- Worked
- 8 min 22 s
- Grade
- Passed
- Score 100 of 100. Graded afterwards.
- How it ended
- Stopped by an error
Every detail of this run
- Agent
- GPT 6 Astra, in Codex
- Benchmark
- Terminal-Bench 4.0
- Asked for
- 4 minutes: “Please work on this task for a full 4 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 8 min 22 s (502 s)
- Ratio
- 2.09× the request
- How it ended
- Stopped by an error
- Grade
- Passed
- It passed the task's tests. The grader returned 1. Pass rule: The task's tests pass.
- Score
- 100 of 100
- Graded afterwards. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- yafmfowoos
- Data release
- 26 Sep 2026
Related runs
| Astra on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest (this run) | 4 minutes | 8 min 22 s | 2.09× longer | |
| Middle | 15 minutes | 15 min 31 s | Within 5% | |
| Longest | 60 minutes | 1 h 1 min | Within 5% | |
| Other agents, shortest request | ||||
| GPT 5.6 Sol | 4 minutes | 21 min 46 s | 5.44× longer | |
| Claude Fable 5.1 | 4 minutes | 12 min 6 s | 3.02× longer | |
See it on the chart. Transcripts are not published yet.
Read before quoting. This run was stopped by an error, so its time is a lower bound: the agent might have kept going. Details
Data and citation
Upstream: Merrill et al., 2026 (ICLR)