Hebbia tested Fable 5.1 on our evaluations, achieving the best fact recall over financial documents of any model tested.
Public source
Publisher name
Public post
On our evals, Fable 5.1 built the best decks and had the best fact recall over financial documents of any model we've tested. It's also the first to fully complete our e…
Company
Hebbia
The leading AI platform for finance.
- Industry
- Software Development
- Location
- New York, US
- Company size
- 51–200 employees
About Hebbia
Hebbia is the leading AI platform for finance. Founded in 2020 by George Sivulka, Hebbia is a generative AI company backed by Andreessen Horowitz, Peter Thiel, and Index Ventures. Investment banks and over 40% of the largest asset managers by AUM use Hebbia’s AI agents to drive investment decisions and automate financial analyst workflows. Users can instantly surface insights over filings, research, and millions of internal documents, enabling citation-backed research, AI-driven document, powerpoint, and spreadsheet generation, and AI driven origination, screening, and diligence. Learn more at hebbia.com.
See moreLatest activity
Latest activity from Hebbia
8 signals
Research & Knowledge
Hebbia evaluated Astra and found it produced the best decks tested and followed the brief 17% more faithfully than the next-best model while sourcing its claims to the right document 19% more often.
Products & Services
Hebbia launched a new SKU that closed several of the biggest contracts in company history within weeks of first customer conversations.
People
Hebbia hired 100 people since January, doubling its headcount across London, San Francisco, and New York City.
Discover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
Athian
Athian switched to the new Fable 5.1 model, which improved from 29% to 52.6% in science test scores, resulting in a 1% accuracy improvement.
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
Samaya AI
Samaya AI released Claude Fable 5.1, a model that scores 55.9% against 49.2% for Fable 5 on FrontierFinance.
Research & Knowledge
Vals AI
Vals AI's GLM 5.3 Flash (codenamed ox-alpha) ranks #5 overall and #1 among open-weight models on Finance Agent v2
Research & Knowledge
Snorkel AI