Pinecone, the vector database startup founded by Edo Liberty, the former head of Amazon’s AI Labs, has long been at the forefront of helping businesses augment large language models (LLMs) with their own data. Most recently, though, the company completely rearchitected its product to launch Pinecone Serverless, which frees its customers from having to think about managing their deployments and scaling them. Today, Pinecone serverless comes out of beta and is now generally available.

Liberty notes that the company’s early customers are now transitioning from experimenting with generative AI to wanting to launch their own AI products. The company watched enterprises grapple with the complexity of building new applications all while also figuring out how to best put them into production.

“The first like wave of production-grade applications is hitting the market now and in the next six to nine months. What our more than 5,000 customers told us loud and clear is that they need a dedicated, optimized, specialized tool that is extremely good at doing vector search, doing RAG, extracting knowledge and generating context for these language models. What they were really saying is: hey, I need scale, I need performance, and I need costs to be such that I can reason about the product that I’m building.”

Liberty stressed that Pinecone spent a lot of time making the product ready for production deployments — all while making it significantly more affordable, too. The company actually believes that customers who use Pinecone serverless can reduce their cost up to 50x, in part because the team rearchitected the system to be a multi-tenant service that decouples storage and compute. With that, Pinecone’s customers only pay when they actually consume CPU time, with the company orchestrating the capacity in the backend.

“Because we run everything as a service, our ability to orchestrate all of that makes us able to charge people for exactly what they use — and not anything more. That is incredibly rare and incredibly hard to do,” Liberty said.

During the public preview, Pinecone’s customers also asked for a number of additional features. One of these is Private Endpoints, which is launching in public preview today. This allows enterprises to create a direct connection to their virtual private clouds on Amazon via AWS PrivateLink, which doesn’t expose their data to the public internet to ensure the data stays well within the various governance and compliance regimes a company may have to adhere to.

Some of the companies that are already using Pinecone serverless include Gong, Help Scout, New Relic, Notion, TaskUS and You.com.

“Notion is leading the AI productivity revolution,” Notion co-founder and COO Akshay Kothari said. “Our launch of a first-to-market AI feature was made possible by Pinecone serverless. Their technology enables our Q&A AI to deliver instant answers to millions of users, sourced from billions of documents. Best of all, our move to their latest architecture has cut our costs by 60%, advancing our mission to make software toolmaking ubiquitous.”

Pinecone, the vector database startup founded by Edo Liberty, the former head of Amazon’s AI Labs, has long been at the forefront of helping businesses augment large language models (LLMs) with their own data. Most recently, though, the company completely rearchitected its product to launch Pinecone Serverless, which frees its customers from having to think about managing their deployments and scaling them. Today, Pinecone serverless comes out of beta and is now generally available.

Liberty notes that the company’s early customers are now transitioning from experimenting with generative AI to wanting to launch their own AI products. The company watched enterprises grapple with the complexity of building new applications all while also figuring out how to best put them into production.

“The first like wave of production-grade applications is hitting the market now and in the next six to nine months. What our more than 5,000 customers told us loud and clear is that they need a dedicated, optimized, specialized tool that is extremely good at doing vector search, doing RAG, extracting knowledge and generating context for these language models. What they were really saying is: hey, I need scale, I need performance, and I need costs to be such that I can reason about the product that I’m building.”

Liberty stressed that Pinecone spent a lot of time making the product ready for production deployments — all while making it significantly more affordable, too. The company actually believes that customers who use Pinecone serverless can reduce their cost up to 50x, in part because the team rearchitected the system to be a multi-tenant service that decouples storage and compute. With that, Pinecone’s customers only pay when they actually consume CPU time, with the company orchestrating the capacity in the backend.

“Because we run everything as a service, our ability to orchestrate all of that makes us able to charge people for exactly what they use — and not anything more. That is incredibly rare and incredibly hard to do,” Liberty said.

During the public preview, Pinecone’s customers also asked for a number of additional features. One of these is Private Endpoints, which is launching in public preview today. This allows enterprises to create a direct connection to their virtual private clouds on Amazon via AWS PrivateLink, which doesn’t expose their data to the public internet to ensure the data stays well within the various governance and compliance regimes a company may have to adhere to.

Some of the companies that are already using Pinecone serverless include Gong, Help Scout, New Relic, Notion, TaskUS and You.com.

“Notion is leading the AI productivity revolution,” Notion co-founder and COO Akshay Kothari said. “Our launch of a first-to-market AI feature was made possible by Pinecone serverless. Their technology enables our Q&A AI to deliver instant answers to millions of users, sourced from billions of documents. Best of all, our move to their latest architecture has cut our costs by 60%, advancing our mission to make software toolmaking ubiquitous.”