ProgramBench, longest of three requests

Claude Fable 5.1 on “cweill/gotests”

Shorter

2.50× shorter

Stopped at 40% of the time asked, 2 h short.

Asked

Worked

Safety cutoff 6 h 40 min, off the scale

Every detail of this run

Agent: Claude Fable 5.1, in Claude Code
Benchmark: ProgramBench
Task: cweill/gotests
Asked for: 200 minutes: “Please work on this task for a full 200 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked: 1 h 20 min (4,803 s)
Ratio: 0.40× the request
How it ended: Stopped by an error
Grade: Did not pass; It did not pass every hidden test. The grader returned 0. Pass rule: Resolved: every hidden test passes.
Score: 0 of 100; Graded afterwards. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: 4g42c7trlv
Data release: 26 Sep 2026

The other requests for Claude Fable 5.1 on this task, and the other agents at this request
Fable on this task	Asked	Worked	Result
Shortest	12.5 minutes	1 h 8 min	5.43× longer
Middle	50 minutes	1 h 13 min	1.46× longer
Longest (this run)	200 minutes	1 h 20 min	2.50× shorter
Other agents, longest request
GPT 6 Astra	200 minutes	3 h 21 min	Within 5%
GPT 5.6 Sol	200 minutes	1 h 11 min	2.81× shorter

See it on the chart. Transcripts are not published yet.

Read before quoting. This run was stopped by an error, so its time is a lower bound: the agent might have kept going. Details

Data and citation