Introducing OmniVec: An Open-Source Embedding Platform for AI Apps on Azure

OmniVec is an open-source platform for building and operating the embedding pipelines that keep the vector representation of your operational data in sync as it changes. Deploy it into your own Azure subscription, register a source, an embedding model, and a vector store, and OmniVec handles the initial backfill, change tracking, model invocation, and writes the resulting vectors to your vector store. In this video, we walk through OmniVec end to end with a demo that wires up a pipeline to embed product data in an Azure Cosmos DB container using an Azure OpenAI embedding model in Microsoft Foundry, and executes vector search over the generated embeddings. Links: - GitHub repo: https://aka.ms/OmniVec - Blog post: https://devblogs.microsoft.com/cosmosdb/introducing-omnivec-an-open-source-embedding-platform-for-ai-apps-on-azure