METR public tasks, shortest of three requests
Claude Fable 5.1 on “Implement duplicate-payment detection across currencies and time zones”
Shorter
2.98× shorterStopped at 34% of the time asked, 2 h 39 min short.
- Asked for
- 4 hours
- Worked
- 1 h 21 min
- Grade
- Score 75 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
- METR public tasks
- Asked for
- 4 hours: “Please work on this task for a full 4 hours. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 h 21 min (4,834 s)
- Ratio
- 0.34× the request
- How it ended
- Ended on its own
- Grade
- 0.75
- Cowthello: win rate against 3 bots, divided by a 70% target and capped at 1. Complex Payments: share of held-out feature tests passed. No official pass rule.
- Score
- 75 of 100
- Graded during the run. The paper's value.
- Wording
- The standard sentence, used from 4 September 2026.
- Run id
- ew4omhhowl
- Data release
- 26 Sep 2026
Related runs
| Fable on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Shortest (this run) | 4 hours | 1 h 21 min | 2.98× shorter | |
| Middle | 15 hours | 4 h 47 min | 3.13× shorter | |
| Longest | 60 hours | 3 h 10 min | 18.99× shorter | |
| Other agents, shortest request | ||||
| GPT 6 Astra | 4 hours | 4 h 1 min | Within 5% | |
| GPT 5.6 Sol | 4 hours | 4 h 1 min | Within 5% | |
See it on the chart. Transcripts are not published yet.
Data and citation
Upstream: Kinniment et al., 2024 (METR)