Crosby tested Fable 5.1 against RedlineBench and saw a model improvement from 47.9 to 57.0
Public source
Publisher name
Public post
At Crosby, we were able to test Fable 5.1 ahead of its launch against RedlineBench and the model improved significantly from 47.9 to 57.0. Internal evaluations are ongoi…
Company
Crosby
The AI Law Firm Built for Execution. Lawyer-assisted AI to help history's fastest companies close deals even faster.
- Industry
- Software Development
- Company size
- 51–200 employees
About Crosby
The AI Law Firm Built for Execution. Lawyer-assisted AI to help history's fastest companies close deals even faster.
See moreDiscover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
Macroscope
Macroscope benchmarked Fable 5.1 alongside other frontier models and measured its highest recall of any model measured so far.
Research & Knowledge
Legora
Legora evaluated performance of GPT-6 Astra using the Legora BAR to compare it to GPT-5.6 Sol, showing clear improvements on output quality across medium and difficult cases.
Research & Knowledge
Hebbia
Hebbia tested Fable 5.1 on our evaluations, achieving the best fact recall over financial documents of any model tested.
Research & Knowledge
Datasite
Datasite released Fable 5.1, which improved pricing and ZDR improvements in every other benchmark over Fable 5, but Financial Reasoning QA scored materially lower.
Research & Knowledge
Athian