Uploaded October 2026 | Updated November 2026, 1 day ago
Add vector search to Open WebUI with LiteLLM and an Oracle AI Database vector store. This walkthrough configures OCI Generative AI embedding models through LiteLLM, connects Open WebUI to that embedding endpoint, and stores the resulting vectors in the database for document search and retrieval-augmented responses.
Chapters:
0:00 Architecture overview: embeddings and vector storage
0:43 Configure OCI credentials and LiteLLM
2:15 Connect Open WebUI to the embedding endpoint
3:19 Configure the database vector store
4:39 Upload documents for testing
5:20 Test vector search across uploaded files
Learn how to configure OCI Generative AI in LiteLLM, use LiteLLM as an embedding proxy for Open WebUI, and connect Open WebUI to a database-backed vector store. The database configuration includes the vector-store setting, wallet-based connectivity, schema owner, DSN or TNS alias, wallet path, TNS_ADMIN path, vector length, and connection-pool settings.
This tutorial is for developers building RAG and document-search experiences. It shows the full path from uploaded documents to embedding generation, persisted vectors, source retrieval, and answers grounded in the uploaded files.
#codeinnovate #VectorSearch #RAG #OpenWebUI #LiteLLM
Add vector search to Open WebUI with LiteLLM and an Oracle AI Database vector store. This walkthrough configures OCI Generative AI embedding models through LiteLLM, connects Open WebUI to that embedding endpoint, and stores the resulting vectors in the database for document search and retrieval-augmented responses.
Chapters:
0:00 Architecture overview: embeddings and vector storage
0:43 Configure OCI credentials and LiteLLM
2:15 Connect Open WebUI to the embedding endpoint
3:19 Configure the database vector store
4:39 Upload documents for testing
5:20 Test vector search across uploaded files
Learn how to configure OCI Generative AI in LiteLLM, use LiteLLM as an embedding proxy for Open WebUI, and connect Open WebUI to a database-backed vector store. The database configuration includes the vector-store setting, wallet-based connectivity, schema owner, DSN or TNS alias, wallet path, TNS_ADMIN path, vector length, and connection-pool settings.
This tutorial is for developers building RAG and document-search experiences. It shows the full path from uploaded documents to embedding generation, persisted vectors, source retrieval, and answers grounded in the uploaded files.
#codeinnovate #VectorSearch #RAG #OpenWebUI #LiteLLM










