Pinecone Blog

In September Pinecone serverless added an evaluation API to evaluate RAG and the ability to choose the LLM, among other updates.

Introducing the ability to bulk import from object storage for Pinecone’s serverless infrastructure. This new capability makes ingesting large amounts of data more efficient for developers building accurate, secure, and scalable AI applications.

Pinecone Assistant is now available in public preview, including new features like a built-in evaluation framework, expanded LLM support, and more.

In August Pinecone serverless became generally available to Google Cloud Platform (GCP) and Microsoft Azure and embedding and reranking were added to the Pinecone API.

Pinecone serverless is generally available (GA) on Microsoft Azure. Along with AWS and GCP, you can now build with serverless on the cloud and region that suits you best.