AppWorld, longest of three requests
GPT 5.6 Sol on “Reconcile Venmo payments into Splitwise with attached receipts”
Within 5%
Within 5%Worked 104% of the time asked, within 5% of it.
- Asked for
- 25 minutes
- Worked
- 25 min 58 s
- Grade
- Passed
- Score 100 of 100. Graded during the run.
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 5.6 Sol, in Codex
- Benchmark
- AppWorld
- Asked for
- 25 minutes: “Please work on this task for a full 25 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 25 min 58 s (1,558 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 during the run. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- ep4vagwitg
- Data release
- 26 Sep 2026
Related runs
| Sol on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest | 1.5 minutes | 1 min 47 s | 1.19× longer | |
| Middle | 6 minutes | 6 min 31 s | 1.09× longer | |
| Longest (this run) | 25 minutes | 25 min 58 s | Within 5% | |
| Other agents, longest request | ||||
| GPT 6 Astra | 25 minutes | 25 min 42 s | Within 5% | |
| Claude Fable 5.1 | 25 minutes | 3 min 3 s | 8.20× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Trivedi et al., 2024 (ACL)