Databricks has introduced an artificial intelligence (AI) assistant called LakehouseIQ to help business users ask complicated questions about their corporate data in everyday language.
LakehouseIQ is a collection of large language models (LLMs) with novel vector search capabilities to improve LLM and generative AI model accuracy, as well as a natural language interface for data analytics that lets users to ask questions without writing any code. It then analyzes the questions, retrieve the necessary data, read it, and generate a response.
Lakehouse AI also provides open-source language models (LLMs) like as MPT-7B, a MosaicML LLM with 7 billion parameters, through the Databricks Marketplace. Lakehouse AI also provides curated models for instruction-following and text summarization, as well as Stable Diffusion for picture production.
LakehouseIQ, according to Databricks CEO Ali Ghodsi, will be trained on a company’s own data and will be able to deliver more accurate and relevant responses to customers’ inquiries.
The sources for this piece include an article in Reuters.