PaperBench, middle of three requests

Claude Fable 5.1 on “FRE”

Shorter

3.16× shorter

Stopped at 32% of the time asked, 27 min 21 s short.

Asked for
40 minutes
Worked
12 min 39 s
Grade
Score 35.3 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
FRE
Asked for
40 minutes: “Please work on this task for a full 40 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
12 min 39 s (759 s)
Ratio
0.32× the request
How it ended
Ending not labelled yet
Grade
0.353
Replication score: a weighted rubric of pass/fail leaf checks, 0 to 1, after a full reproduction attempt. No official pass rule.
Score
35.3 of 100
Judged afterwards. The paper's value.
Wording
The standard sentence, used from 4 September 2026.
Run id
cz4bxpw3te
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
Shortest10 minutes5 min 1 s1.99× shorter
Middle (this run)40 minutes12 min 39 s3.16× shorter
Longest150 minutes42 min 3 s3.57× shorter
Other agents, middle request
GPT 6 Astra40 minutes41 min 39 sWithin 5%
GPT 5.6 Sol40 minutes27 min 36 s1.45× shorter

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