It has an end-to-end platform that accelerates AI application development by 100x across the entire AI application lifecycle, including data preparation and preparation, application development, real-time data, and deployment.

Applications of AI


DataStax enables developers to take advantage of several new updates to focus on application development instead of infrastructure management.

Langflow 1.0 and DataStax Langflow release
In April, DataStax acquired Langflow, a popular open source visual framework for building RAG applications. Now, DataStax has released Langflow 1.0, which includes a version of Langflow hosted on the DataStax Cloud platform.

Langflow 1.0's drag-and-drop interface features integration with many of the leading Gen AI tools, including LangChain, LangSmith, OpenAI, Hugging Face, and Mistral, allowing developers to easily set up, swap, and compare leading large-scale language model and embedding providers.




This gives developers great flexibility to easily compare different providers and their results. Developers can make major changes in just a few minutes without having to learn new APIs and recode their applications.

Additionally, as part of the Langflow 1.0 open source release, developers can now leverage LangSmith’s observability services to trace application responses and create more relevant and accurate LLM-based applications.

RAG-Ready Your Data with Unstructured.io
The new partnership between DataStax and Unstructured makes it easy for enterprises and developers to AI-enable their enterprise data by handling data ingestion and chunking across data types such as PDF, Salesforce, Google Drive, and more for use in AI applications.

Developers can benefit from ultra-fast data ingestion by quickly converting large datasets and common document types into vector data. This new integration enables these embeddings to be quickly written to Astra DB for highly relevant genAI similarity searches. Additionally, for those managing very large datasets, users can convert that data into embeddings and write them to Astra DB in just minutes.

DataStax Vectorise lets you leverage the largest ecosystem of embedding providers in minutes.
Vectorise simplifies vector generation by allowing developers to choose an embedding service, configure it in Astra DB, and start building right away. Today, most embedding is handled “client-side”, forcing developers to learn different APIs. With DataStax Vectorise, vector embedding happens on the server, so developers only need to learn one API to access the eight most popular embedding providers and compare their results. DataStax partners with top embedding providers to offer a robust selection and simplify implementation so users only need to configure one API to create embeddings. Partner embedding providers include Azure OpenAI, Hugging Face, Jina AI, Mistral AI, Nvidia, OpenAI, Upstage AI, and Voyage AI.

Combining the best of genAI open source with RAGStack 1.0
Since the release of RAGStack in December 2023, DataStax has continued to expand the depth and breadth of the product by adding several key capabilities, integrations, and partnerships, all of which are now available in the RAGStack 1.0 release, a production-ready, out-of-the-box solution with an efficient set of tools, techniques, and governance to streamline RAG implementations at enterprise scale.

Today, every company building with genAI is looking for the most effective way to implement RAGs within their applications. Companies need a proven way to succeed with genAI. Companies rely on external APIs that come with no guarantees, are released on their own schedules, and often threaten the stability of the applications they deliver. Companies cannot rely on unsupported open source projects or vendors that cannot effectively support the needs and scale of genAI projects.

The RAGStack 1.0 release provides stability for all genAI applications and frameworks by delivering the best of open source and the latest technology required for enterprise use cases. RAGStack 1.0 includes several new features:

· Langflow in RAGStack – Users can build applications faster with Langflow using RAGStack versions of components that have been 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.

· RAGStack's ColBERT with Astra DB – First production-ready implementation leveraging Astra DB to achieve significantly better repeatability than single-vector encoding.

· Introduce Text2SQL/Text2CQL to bring structured, semi-structured and unstructured context into genAI flows, activating existing data for further benefits.

Vectorization with RAGStack and LangChain – Allows open source frameworks to leverage new server-side embedding patterns using chains.

“The generative AI stack is a large, complex chunk of technology that many are trying to understand. We're focused on enabling developers to stay true to their roots and focus on what they do best: building and developing, without having to worry about application infrastructure. We're providing a cutting-edge, end-to-end stack to make this that much easier,” said Ed Anuff, chief product officer at DataStax. “From releasing Langflow in Astra to bringing together the largest ecosystem of built-in providers in one place, we're delivering on our promise to make genAI application development as fast and easy as possible, enabling organizations to get apps into production for immediate impact.”

“Mistral AI paves the way for developers to generate embeddings seamlessly,” said Sophia Yang Ph.D., head of developer relations at Mistral AI. “Our collaboration with DataStax accelerates development, allowing developers to focus on improving core functionality while streamlining the complexity of embedding models.”

“By partnering with DataStax, Unstructured is giving developers the tools to seamlessly extract and transform complex data, then store it in Astra DB Vector to power LLM-based applications,” said Brian Raymond, founder and CEO of Unstructured. “This partnership will significantly enhance genAI applications by speeding up data ingest, reducing compute overhead, and improving scalability.”

“Upstage is pleased to partner with DataStax to deliver unparalleled performance and cost efficiency by running a full-stack LLM solution within Astra Vectorise,” said Sung Kim, CEO and co-founder of Upstage. “By partnering with DataStax, we can provide developers with a solution that delivers tangible results while abstracting the complexities of embedding models from application code.”

“Vectorise gives developers access to advanced AI capabilities never before possible,” said Saahil Ognawala, head of product at Jina AI. “By integrating Jina Embeddings into Vectorise, DataStax is simplifying the development process, allowing developers to focus on improving core functionality without the hassle of integrating external systems.”

“Seamlessly integrated into Astra Vectorise, our domain-customized embedded models abstract the complexity of embedded models from application code, improving search quality and optimizing workflows,” said Tengyu Ma, CEO of Voyage AI. “Through this partnership, we will empower developers to achieve meaningful application outcomes through an enhanced developer experience.”

About DataStax
DataStax is the company that empowers developers and enterprises to successfully build a bold new world through genAI. We provide a one-stop generative AI stack with everything you need to put relevant, responsive GenAI applications into production faster and easier. DataStax delivers a RAG-first developer experience with first-class integrations with leading AI ecosystem partners and works with your existing stack of choice. DataStax enables anyone to rapidly build smart, fast-growing AI applications at unlimited scale on any cloud. Hundreds of the world's leading companies rely on DataStax, including Audi, Bud Financial, Capital One, and Skypoint. Learn more at DataStax.com.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *