PaperBench, longest of three requests

Claude Fable 5.1 on “BAM”

Shorter

3.63× shorter

Stopped at 28% of the time asked, 58 min short.

Asked for
80 minutes
Worked
22 min 1 s
Grade
Score 67.1 of 100
Judged afterwards.
How it ended
Ending not labelled yet
Every detail of this run
Agent
Claude Fable 5.1, in Claude Code
Benchmark
PaperBench
Task
BAM
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
22 min 1 s (1,321 s)
Ratio
0.28× the request
How it ended
Ending not labelled yet
Grade
0.671
Replication score: a weighted rubric of pass/fail leaf checks, 0 to 1, after a full reproduction attempt. No official pass rule.
Score
67.1 of 100
Judged afterwards. The paper's value.
Wording
The standard sentence, used from 4 September 2026.
Run id
3mtjfhr4gu
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 taskAskedWorkedSplitResult
Shortest5 minutes6 min 14 s1.25× longer
Middle20 minutes11 min 17 s1.77× shorter
Longest (this run)80 minutes22 min 1 s3.63× shorter
Other agents, longest request
GPT 6 Astra80 minutes1 h 21 minWithin 5%
GPT 5.6 Sol80 minutes1 h 31 min1.14× longer

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

Read before quoting. How this run ended has not been labelled yet. It is timed like every other run and counted in every number. Details

Data and citation

Upstream: Starace et al., 2025