Vector DB & Data, get retrieval right

AI vector database and embedding tool for semantic search and RAG applications

Fully managed vector database with no operations work; the Starter tier is free forever with 2GB of storage, and reads, writes and storage are billed separately

Open-source vector database under BSD-3 that is free to self-host; its focus is hybrid retrieval that blends vector similarity with BM25 keyword search

CNCF open-source vector database under Apache 2.0, built for billion-scale retrieval; since 3.0 it can index data directly on lakehouse formats such as Iceberg and Lance

Open-source vector database written in Rust under Apache 2.0; the cloud offers a permanent free tier, while Standard bills by vCPU and memory with no separate query or vector charges

Open-source retrieval infrastructure for AI under Apache 2.0, putting vector, full-text and metadata search behind one interface and usable as an embedded library

SoMark's open-source 1.3B document parser released in late October 2026, outputting Markdown or JSON with layout coordinates and reading order, ranked first on olmOCR-Bench at 87.84, OmniDocBench at 97.18 and DocParsingBench at 87.76, with a replaceable layout module

OpenBMB's Apache 2.0 document recognition suite released 2026/10/27, pairing a 1.3B parser with the 2B Jev quality model that scores text, formulas and tables from 0 to 1, with 93.68 on Real5-OmniDocBench

LlamaIndex's draft answer key for document parsing published 2026/10/28, with 140 rules across text, tables, forms, charts and diagrams, each paired with an example page and expected output

Enterprise reranker, Pro and Fast tiers, passage-level Context Focus

Open robot video labeling suite: segment, caption and score training data

Open-source vector index from UC Berkeley's Sky Computing Lab and best paper at MLSys 2026, storing only a pruned graph and recomputing embeddings at query time so a 60 million chunk Wikipedia index falls from 201GB to 6GB, MIT licensed