Pinecone Changelog: Hybrid search in one index, BYOC, and VQ-Bench

Hybrid search in Pinecone now runs from one index: dense vectors, sparse vectors, full-text search, and metadata, with integrated inference. Plus: BYOC is generally available, VQ-Bench for vector quantizers, and a cost and usage export API. What's new: Documents API, now with integrated inference: run dense, sparse, full-text, and metadata search from one index, and let Pinecone embed your text fields for you. BYOC is generally available on AWS, Google Cloud, and Azure, and now supports on-demand capacity. Your data stays in your cloud account while Pinecone runs the operations. VQ-Benc