Pinecone’s vector database gets a new serverless architecture

Pinecone’s vector database gets a new serverless architecture

Pinecone’s vector database gets a new serverless architecture

Frederic Lardinois@fredericl / 9:00 PM GMT+7•January 16, 2024

 Comment

Image Credits: D-BASE / Getty Images

For a long time, vector databases were a bit of a niche product, but because they are uniquely suited to provide context and long-term memory to large language models, everybody in the database space is now seemingly trying to bolt vector search onto their existing products as fast as possible. Meanwhile, dedicated services like Pinecone, which was founded by the team behind Amazon SageMaker, are leading the charge, with Pinecone raising a total of $138 million since it as founded in 2019. Today, Pinecone is launching Pinecone Serverless, a new and significantly enhanced serverless architecture to power its service.

Pinecone Serverless now separates reads, writes and storage, which should reduce costs for users. Indeed, Pinecone argues that its new architecture can offer a 10x to 100x cost reduction. The new architecture now supports vector clustering on top of blob storage. This results in lower latencies and the ability for Pinecone Serverless to support massive data sizes. Likewise, Pinecone Serverless introduces new indexing and retrieval algorithms to enable fast vector search across this blob storage. The service now also offers a multi-tenant compute layer.

“Since it is truly serverless, it completely eliminates the need for developers to provision or manage infrastructure and allows them to build GenAI applications more easily and bring them to market much faster,” the company explains in its announcement. “As a result, developers with use cases of any size can build more reliable, effective, and impactful GenAI applications with any LLM of their choice, leading to an imminent wave of incredible GenAI applications reaching the market.”

From the outset, Pinecone Serverless will offer integrations with several other AI and back-end services, including Anthropic, Anyscale, Cohere, Confluent, LangChain, Pulumi and Vercel. “Vercel’s mission is to help the world ship the best products, and in the age of GenAI that requires Pinecone as the vector database component,” said Guillermo Rauch, CEO and founder of Vercel. “That’s why we are announcing that all Vercel users can now add Pinecone Serverless to their applications in just a few clicks, with more exciting capabilities to come.”