Terminal-Bench 4.0, longest of three requests
Claude Fable 5.1 on “Implement a synthesizable 8-bit retro console system-on-chip”
Shorter
4.19× shorterStopped at 24% of the time asked, 3 h 10 min short.
- Asked for
- 250 minutes
- Worked
- 59 min 41 s
- Grade
- Passed
- Score 100 of 100. Graded during the run.
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Claude Fable 5.1, in Claude Code
- Benchmark
- Terminal-Bench 4.0
- Asked for
- 250 minutes: “Please work on this task for a full 250 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 59 min 41 s (3,581 s)
- Ratio
- 0.24× the request
- How it ended
- Ended on its own
- Grade
- Passed
- It passed the task's tests. The grader returned 1. Pass rule: The task's tests pass.
- Score
- 100 of 100
- Graded during the run. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- 6sujjzv5r7
- Data release
- 26 Sep 2026
Related runs
| Fable on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest | 15 minutes | 15 min 57 s | 1.06× longer | |
| Middle | 60 minutes | 20 min 51 s | 2.88× shorter | |
| Longest (this run) | 250 minutes | 59 min 41 s | 4.19× shorter | |
| Other agents, longest request | ||||
| GPT 6 Astra | 250 minutes | 4 h 23 min | 1.05× longer | |
| GPT 5.6 Sol | 250 minutes | 4 h 11 min | Within 5% | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Merrill et al., 2026 (ICLR)