Langflow Updates and More: DataStax Launches Massive New Updates at Rag++ Event
Featuring End-to-End Platform That Makes AI Application Development 100x Faster
DataStax, the Artificial Intelligence (AI) platform company, has announced major updates to its Generative AI development platform that help make retrieval augmented generation (RAG) powered application development 100X faster. DataStax showcased its newly released updates at the RAG++ event in San Francisco with partners such as LangChain, Microsoft, NVIDIA, and Unstructured, among others.
With DataStax, developers can focus on application development, rather than infrastructure management, powered by multiple, new updates:
Launching Langflow 1.0 and DataStax Langflow
In April, DataStax acquired Langflow, the popular, open source visual framework for building RAG applications. Now, DataStax is releasing Langflow 1.0, which includes a version of Langflow hosted in the DataStax Cloud platform.
Langflow 1.0âs drag and drop interface, with dozens of integrations with the top Gen AI toolsâLangChain, LangSmith, OpenAI, Hugging Face, Mistral, and othersâmakes it easy for developers to set up, swap, and compare all the major large language model and embedding providers.
This gives developers tremendous flexibility to easily compare different providers and their results. Developers can now make major changes in just minutes instead of having to learn new APIs and re-coding their applications.
âThe Generative AI stack is a big and complex ball of technology that many are working to get their arms around. Weâre focused on helping developers stay true to their roots so they can do what they do best: build and develop, rather than worrying about application infrastructure. Weâre delivering a cutting-edge, end-to-end stack to make this a lot easier,â said Ed Anuff, Chief Product Officer at DataStax. âFrom the launch of Langflow in Astra, to bringing the largest ecosystem of embedding providers together in one place, weâre delivering on our promise to make GenAI application development as fast and simple as possible so organizations can get their apps in production quickly, for immediate impact.â
Additionally, as part of the Langflow 1.0 open source release, developers can now leverage LangSmithâs observability service to trace an applicationâs responses for more relevant, accurate LLM-based applications.
For more information, read the DataStax Langflow launch blog post.
Making Data RAG-ready with Unstructured.io
A new partnership between DataStax and Unstructured enables enterprises and developers to easily make their enterprise data ready for AI, handling the data ingestion and chunking across data types: PDFs, Salesforce, Google Drive, etc. to use in AI applications.
Developers benefit from lightning-fast data ingestion through quick conversion of large data sets and common document types into vector data. This new integration then enables these embeddings to be quickly written to Astra DB for highly relevant GenAI similarity searches. And, when managing very large datasets, users are able to convert that data into embeddings and write them to Astra DB in just minutes.
“Partnering with DataStax, Unstructured equips developers with the tools to seamlessly extract and transform complex data, storing it in Astra DB Vector to power LLM-based applications,â said Brian Raymond, Founder of and CEO at Unstructured. âThis partnership significantly enhances GenAI applications by delivering faster data retrieval, reducing computational overhead, and boosting scalability.”
Read more about the Unstructured partnership.
Leverage the Largest Ecosystem of Embedding Providers in Minutes, with DataStax Vectorize
Vectorize simplifies vector generation by letting developers choose an embedding service, configure it with Astra DB, and start building right away. Most embedding is currently handled âclient-sideâ- meaning that Developers need to learn many different APIs.
With DataStax Vectorize, vector embedding now happens on the server; meaning developers only need to learn one API to now access the eight most popular embedding providers and compare results between them. DataStax has partnered with the top embedding providers to offer robust choice, with simplified implementation that only requires users to configure a single API to create embeddings. Partner embedding providers include: Azure OpenAI, Hugging Face, Jina AI, Mistral AI, NVIDIA, OpenAI, Upstage AI, and Voyage AI.
Read more about DataStax Vectorize.
âUpstage is thrilled to partner with DataStax to offer unparalleled performance and cost-effectiveness of running our full-stack LLM solution within Astra Vectorize,â said Sung Kim, Co-Founder of and CEO at Upstage. âOur partnership with DataStax enables us to provide developers with solutions that drive tangible results, all while abstracting the complexity of embedding models from the application code.â
âWith Vectorize, developers gain unprecedented access to advanced AI capabilities,â said Saahil Ognawala, Head of Product at Jina AI. âBy integrating Jina Embeddings within Vectorize, DataStax simplifies the development journey, empowering developers to focus on refining their core functionalities without the hassle of external system integrations.”
Bringing the Best of GenAI Open Source Together with RAGStack 1.0
Since the launch of RAGStack in December 2023, DataStax has added continued depth and breadth to the product with the additions of several key features, integrations, and partnerships, all available now in the RAGStack 1.0 releaseâthe production-ready, out-of-the-box solution that streamlines RAG implementation at enterprise scale, with an efficient set of tools, techniques, and governance.
Every company building with GenAI right now is looking for the most effective way to implement RAG within their applications. Enterprises need proven paths to success with GenAI. They are dependent on external APIs that have no guarantees, release on their own schedule, and often threaten the stability of the applications theyâre serving. Enterprises canât depend on unsupported open source projects or vendors who cannot effectively support the needs and scale of their GenAI projects.
RAGStackâs 1.0 release provides stability to all GenAI applications and frameworks by offering the best of open source and the latest techniques required for enterprise use cases. RAGStack 1.0 includes multiple new features:
- Langflow in RAGStack. Users can build applications faster with Langflow using RAGStackâs version of components tested for compatibility, performance and security.
- Knowledge Graph RAG. Provides a graph-based representation designed specifically for GenAI applications to store and retrieve information more efficiently and accurately than vector-based similarity search alone with Astra DB.
- ColBERT in RAGStack with Astra DB. The first production-ready implementation delivering significantly better recall than any single-vector encodings, backed by Astra DB.
- Text2SQL/Text2CQL. Brings structured, semistructured, and unstructured context into the GenAI flow activating existing data with additional benefits.
- Vectorize in RAGStack and LangChain. Enables the open source frameworks to leverage a new server-side embedding pattern with chains.