Note
For pgvector demo, check out the pgvector branch.
A sample app showing how to use Ollama, Microsoft.Extensions.AI and OpenAI to generate vector embeddings and implement semantic search in .NET.
Vector embeddings can be generated locally using OllamaSharp or via the OpenAI API. The generated embeddings are stored in a SQL Server database and queried with Entity Framework Core and EFCore.SqlServer.VectorSearch
General:
- The Illustrated Word2vec
- Vector Embeddings Explained
- Distance Metrics in Vector Search
- Why is Vector Search so fast?
- Using Vector Databases for Multimodal Embeddings and Search - Zain Hasan - NDC London 2024
PostgreSQL specific:
- Vectors are the new JSON in PostgreSQL
- Vectors are the new JSON (PGConf.EU 2023 Recording)
- Postgres is all you need, even for vectors
- Vector Indexes in Postgres using pgvector: IVFFlat vs HNSW
- Understanding vector search and HNSW index with pgvector
SQL Server specific: