TUA-Bench, longest of three requests
GPT 6 Astra on “APA references review”
Within 5%
Within 5%Worked 101% of the time asked, within 5% of it.
- Asked for
- 40 minutes
- Worked
- 40 min 28 s
- Grade
- Did not pass
- Score 52.5 of 100. Graded during the run.
- How it ended
- Ended on its own
Every detail of this run
- Agent
- GPT 6 Astra, in Codex
- Benchmark
- TUA-Bench
- Asked for
- 40 minutes: “Please work on this task for a full 40 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 40 min 28 s (2,428 s)
- Ratio
- 1.01× the request
- How it ended
- Ended on its own
- Grade
- Did not pass
- It did not complete the task in full. The grader returned 0.525. Pass rule: Full completion: a reward of 1 on the final state.
- Score
- 52.5 of 100
- Graded during the run. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- wtspodzolb
- Data release
- 26 Sep 2026
Related runs
| Astra on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest | 2.5 minutes | 2 min 45 s | 1.10× longer | |
| Middle | 10 minutes | 10 min 8 s | Within 5% | |
| Longest (this run) | 40 minutes | 40 min 28 s | Within 5% | |
| Other agents, longest request | ||||
| GPT 5.6 Sol | 40 minutes | 41 min 5 s | Within 5% | |
| Claude Fable 5.1 | 40 minutes | 9 min 15 s | 4.32× shorter | |
| Muse Spark 1.3 | 40 minutes | 2 min 51 s | 14.02× shorter | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Chen et al., 2026