Guides
Practical how-to content for engineers building with vector search — RAG pipelines, hybrid search, namespace branching, object store setup, and embedding model migrations.
Migrating from Pinecone to ManyVector
Step-by-step migration guide: export vectors from Pinecone, import into ManyVector, verify recall parity, and cut over traffic without downtime.
Hybrid search query patterns
Practical patterns for combining vector and full-text search with ManyVector: alpha tuning, pre-filtering, re-ranking, and corpus-specific strategies.
Namespace branching for A/B testing embedding models
How to use ManyVector's copy-on-write namespace forks to safely test new embedding models in production without disrupting live traffic.
Building a RAG pipeline with ManyVector
End-to-end guide: chunk documents, generate embeddings, store in ManyVector, and retrieve grounding context for LLM calls. With hybrid search for higher recall.
Connecting your S3 bucket
How to configure AWS S3 as your ManyVector object store backend — IAM policies, bucket setup, encryption, and access logging.
Getting started: upsert your first vectors
A step-by-step guide to creating a ManyVector namespace, connecting your object store, and upserting your first embeddings in under 10 minutes.