Embeddings LiteLLM icon

Embeddings LiteLLM

For advanced usage with an AI chain

Embeddings LiteLLM

AI-generated

Summary

Generate embeddings for input data using a specified LiteLLM embedding model connected via the LiteLLM credentials.

Inputs

  • model — Select the embedding model to generate embeddings, filtered to embedding models available from the configured service.
  • options.batchSize — Maximum number of documents to send in each request (default 512, max 2048).
  • options.stripNewLines — Whether to strip new lines from the input text before generating embeddings (default true).
  • options.timeout — Maximum time in seconds allowed for the embedding request; -1 means no timeout (default -1).

Output shape

a single AI embedding result per execution

The operation returns embeddings generated by the specified model. The node must be connected to a vector store input to function properly.

Examples

Example 1: Generate vector embeddings from input text or documents for use in AI workflows or vector stores.

Choose the embedding model, optionally adjust batch size, newline stripping, and timeout. Connect the node to a compatible vector store to pass embeddings downstream.

Links

Discussion