Terminal-Bench 4.0, shortest of three requests
GPT 5.6 Sol on “Repair batched LLM evaluation so it matches single-example semantics”
Longer
4.08× longerWorked 308% past the request, 24 min 41 s over.
- Asked for
- 8 minutes
- Worked
- 32 min 41 s
- Grade
- Did not pass
- Score 0 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
- Terminal-Bench 4.0
- Asked for
- 8 minutes: “Please work on this task for a full 8 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 32 min 41 s (1,961 s)
- Ratio
- 4.08× the request
- How it ended
- Ended on its own
- Grade
- Did not pass
- It did not pass the task's tests. The grader returned 0. Pass rule: The task's tests pass.
- Score
- 0 of 100
- Graded during the run. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- asv7fmcfmq
- Data release
- 26 Sep 2026
Related runs
| Sol on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest (this run) | 8 minutes | 32 min 41 s | 4.08× longer | |
| Middle | 30 minutes | 30 min 24 s | Within 5% | |
| Longest | 125 minutes | 2 h 6 min | Within 5% | |
| Other agents, shortest request | ||||
| GPT 6 Astra | 8 minutes | 8 min 11 s | Within 5% | |
| Claude Fable 5.1 | 8 minutes | 16 min 30 s | 2.06× longer | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Merrill et al., 2026 (ICLR)