PaperBench, longest of three requests

Claude Fable 5.1 on “Bridging Data Gaps”

Shorter

4.40× shorter

Stopped at 23% of the time asked, 2 h 35 min short.

Asked

Worked

Safety cutoff 6 h 40 min, off the scale

Asked for: 200 minutes
Worked: 45 min 29 s
Grade: Grading…; The run's output is saved, but grading it needs GPU time for the benchmark's judge or evaluation, which has not run yet.
How it ended: Stopped by an error

Every detail of this run

Agent: Claude Fable 5.1, in Claude Code
Benchmark: PaperBench
Task: Bridging Data Gaps
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: 45 min 29 s (2,729 s)
Ratio: 0.23× the request
How it ended: Stopped by an error
Grade: Grading…; The run's output is saved, but grading it needs GPU time for the benchmark's judge or evaluation, which has not run yet.
Wording: The standard sentence, used from 4 September 2026.
Run id: eo57s5gcg5
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	5 min 18 s	2.36× shorter
Middle	50 minutes	15 min 13 s	3.29× shorter
Longest (this run)	200 minutes	45 min 29 s	4.40× shorter
Other agents, longest request
GPT 6 Astra	200 minutes	3 h 21 min	Within 5%
GPT 5.6 Sol	200 minutes	3 h 20 min	Within 5%

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