AssistantBench, longest of three requests
GPT 6 Astra on “AssistantBench validation task 7”
Within 5%
Within 5%Worked 102% of the time asked, within 5% of it.
- Asked for
- 20 minutes
- Worked
- 20 min 25 s
- Grade
- Did not pass
- Score 11.5 of 100. Graded afterwards.
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6 Astra, in Codex
- Benchmark
- AssistantBench
- Asked for
- 20 minutes: “Please work on this task for a full 20 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 20 min 25 s (1,225 s)
- Ratio
- 1.02× the request
- How it ended
- Ended on its own
- Grade
- Did not pass
- It did not match the reference answer exactly. The grader returned 0.115. Pass rule: Exact match with the reference answer.
- Score
- 11.5 of 100
- Graded afterwards. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- zwpn3xkhy4
- Data release
- 26 Sep 2026
Related runs
| Astra on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest | 1.25 minutes | 1 min 40 s | 1.33× longer | |
| Middle | 5 minutes | 8 min 18 s | 1.66× longer | |
| Longest (this run) | 20 minutes | 20 min 25 s | Within 5% | |
| Other agents, longest request | ||||
| GPT 5.6 Sol | 20 minutes | 18 min 45 s | 1.07× shorter | |
| Claude Fable 5.1 | 20 minutes | 11 min 47 s | 1.70× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Yoran et al., 2024 (EMNLP)