ProgramBench, shortest of three requests
GPT 6 Astra on “cweill/gotests”
Within 5%
Within 5%Worked 103% of the time asked, within 5% of it.
- Asked for
- 12.5 minutes
- Worked
- 12 min 53 s
- Grade
- Did not pass
- Score 64.5 of 100. Graded afterwards.
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6 Astra, in Codex
- Benchmark
- ProgramBench
- Task
- cweill/gotests
- Asked for
- 12.5 minutes: “Please work on this task for a full 12.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 12 min 53 s (773 s)
- Ratio
- 1.03× the request
- How it ended
- Ended on its own
- Grade
- Did not pass
- It did not pass every hidden test. The grader returned 0.645. Pass rule: Resolved: every hidden test passes.
- Score
- 64.5 of 100
- Graded afterwards. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- xw2ro5i3i6
- Data release
- 26 Sep 2026
Related runs
| Astra on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest (this run) | 12.5 minutes | 12 min 53 s | Within 5% | |
| Middle | 50 minutes | 50 min 38 s | Within 5% | |
| Longest | 200 minutes | 3 h 21 min | Within 5% | |
| Other agents, shortest request | ||||
| GPT 5.6 Sol | 12.5 minutes | 19 min 48 s | 1.58× longer | |
| Claude Fable 5.1 | 12.5 minutes | 1 h 8 min | 5.43× longer | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yang et al., 2026