AppWorld, longest of three requests
GPT 6 Astra on “Rebuild and send a drafted recommendation email from the user’s Spotify history”
Within 5%
Within 5%Worked 104% of the time asked, within 5% of it.
- Asked for
- 30 minutes
- Worked
- 31 min 11 s
- Grade
- Passed
- Score 100 of 100. Graded afterwards.
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6 Astra, in Codex
- Benchmark
- AppWorld
- Asked for
- 30 minutes: “Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 31 min 11 s (1,871 s)
- Ratio
- 1.04× the request
- How it ended
- Ended on its own
- Grade
- Passed
- It passed every final-state test. The grader returned 1. Pass rule: Every final-state test passes.
- Score
- 100 of 100
- Graded afterwards. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- ljfe7rim3w
- Data release
- 26 Sep 2026
Related runs
| Astra on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest | 2 minutes | 2 min 24 s | 1.20× longer | |
| Middle | 8 minutes | 8 min 21 s | Within 5% | |
| Longest (this run) | 30 minutes | 31 min 11 s | Within 5% | |
| Other agents, longest request | ||||
| GPT 5.6 Sol | 30 minutes | 31 min 34 s | 1.05× longer | |
| Claude Fable 5.1 | 30 minutes | 4 min 26 s | 6.77× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Trivedi et al., 2024 (ACL)