DataStax has announced the general availability of its Data API, a one-stop API for GenAI. DataStax also announced an updated developer experience for Astra DB, DataStax's NoSQL cloud database that is built on Apache Cassandra.

The DataStax Data API service is designed to provide all the data and a complete stack for production GenAI and retrieval augmented generation (RAG) applications with high relevancy and low latency. It will make it easier for developers to work around Gen AI applications, particularly for JavaScript developers that will now be able to integrate vector search and LLMs into their applications.

The new vector Data API makes Cassandra's petabyte-scale power available to JavaScript and Python application developers who are working in AI development. DataStax say the API is designed for ease of use, while offering up to 20% higher relevancy, 9x higher throughput, and up to 74x faster response times than Pinecone. Behind the scenes this is achieved using the JVector search engine.

JVector is a pure Java embedded vector search engine, used by DataStax Astra DB. The JVector developers say it uses state of the art graph algorithms inspired by DiskANN and related research that offer high recall and low latency. The use of JVector provides higher relevance, throughput and volume compared to other vector databases using HNSW (Hierarchical Navigable Small World), according to the DataStax team.

DataStax says the new service also introduces an intuitive dashboard, efficient data loading and exploration tools, and seamless integration with leading AI and machine learning (ML) frameworks.

The API also provides a single stack for Retrieval Augmented Generation to make answers more accurate. It gives users the ability to query data while also indexing new data updates.

The Data API provides integrations with LangChain, OpenAI, Vercel, GCP Vertex, AWS, Azure, Amazon Bedrock and GitHub Copilot. The DataStax team says it means any developer can now support advanced RAG techniques such as FLARE and ReAct that must synthesize multiple responses.

More Information

Astra DB

JVector On GitHub

Related Articles

DataStax Astra DB gets Change Data Capture

DataStax Extends Stargate

DataStax Adds gRPC To Stargate For Performant Microservices

Cassandra 4.1 Focuses On Pluggability

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info