by alvivanco on 4/8/23, 7:05 PM with 1 comments
by alvivanco on 4/8/23, 7:07 PM
The model uses proprietary and curated datasets, including financial news, company financial filings, and press releases.
The Bloomberg data used for training spans from March 1, 2007 through July 31, 2022.
Training for the BloombergGPT model required approximately 53 days of computations run on 64 servers
Use cases include:
1/ generating an initial draft of an SEC filing
2/ summarizing a blurb containing financial content into a headline
3/ providing a company chart of an organization and linkages between an individual and multiple companies
4/ automation of generation of draft routine market reports and summaries for clients
5/ retrieval of specific elements of financial statements for specific periods via a single prompt