Oracle has launched HeatWave GenAI, featuring in-database large language models (LLMs) and an automated vector store, making it easier for customers to build generative AI applications without AI expertise or additional costs. This new offering is available in all Oracle Cloud regions and simplifies the creation of a vector store for enterprise content with a single SQL command.
HeatWave GenAI introduces innovative features, such as LLM in Database for easier development of generative AI applications, automated vector storage in the database for seamless use of AI with business documents, scalable vector processing for fast semantic search results, and Talk about heat waves, a Visual Code plugin for MySQL Shell.
Performance benchmarks show that HeatWave GenAI is significantly faster and more cost-effective compared to other platforms like Amazon Bedrock, Snowflake, Databricks, and Google BigQuery. Additionally, HeatWave’s similarity search processing outperforms Amazon Aurora PostgreSQL with pgvector.
Customers like SmarterD and EatEasy have praised HeatWave GenAI for its ease of use, reduced application complexity, and improved performance. Analysts like Holger Mueller of Constellation Research have commended HeatWave for its integrated approach to generative AI and cost-effective performance advantages.
HeatWave is the only cloud service that offers automated generative AI and machine learning for lake-scale transactions and analytics. It is available on multiple cloud platforms and in customer data centers with OCI Dedicated Region and Oracle Alloy.
Article Source
https://blockchain.news/news/oracle-introduces-in-database-llms-automated-vector-store-heatwave-genai