Package index • ragnar

Document processing

Ingest documents from the web, PDFs, Word, or local Markdown.

Chunk and augment text context.

Store embeddings in DuckDB for fast, local queries.

ragnar_store_create() ragnar_store_connect(): Create and connect to a vector store
ragnar_store_insert() ragnar_store_update(): Inserts or updates chunks in a RagnarStore
ragnar_store_ingest(): Concurrently ingest documents into a Ragnar store
ragnar_store_build_index(): Build a Ragnar Store index
ragnar_store_inspect(): Launch the Ragnar Store Inspector
ragnar_store_atlas(): Visualize a store using Embedding Atlas

Embed chunks with commercial or open-source models.

embed_azure_openai(): Uses Azure AI Foundry to create embeddings
embed_bedrock(): Embed text using a Bedrock model
embed_databricks(): Embed text using a Databricks model
embed_google_gemini() embed_google_vertex(): Embed using Google Vertex API platform
embed_ollama() embed_openai() embed_lm_studio(): Embed Text
embed_snowflake(): Generate embeddings using Snowflake

Retrieve relevant chunks using both vector and text search.