DeepSWE v1.1, longest of three requests
Claude Fable 5.1 on “Add duration-aware sharding to Vitest”
Shorter
2.87× shorterStopped at 35% of the time asked, 52 min 10 s short.
- Asked for
- 80 minutes
- Worked
- 27 min 50 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
- DeepSWE v1.1
- Asked for
- 80 minutes: “Please work on this task for a full 80 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 27 min 50 s (1,670 s)
- Ratio
- 0.35× the request
- How it ended
- Ended on its own
- Grade
- Passed
- It passed its hidden tests. The grader returned 1. Pass rule: The hidden 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
- zxg7ryfttb
- Data release
- 26 Sep 2026
Related runs
| Fable on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest | 5 minutes | 16 min 14 s | 3.25× longer | |
| Middle | 20 minutes | 20 min 7 s | Within 5% | |
| Longest (this run) | 80 minutes | 27 min 50 s | 2.87× shorter | |
| Other agents, longest request | ||||
| GPT 6 Astra | 80 minutes | 1 h 21 min | Within 5% | |
| GPT 5.6 Sol | 80 minutes | 1 h 21 min | Within 5% | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Huang et al., 2026