Embeddings LM Studio SupabaseInstall

Summary

Generate vector embeddings from text inputs using a selected LM Studio embedding model for integration with AI workflows involving vector stores, retrievers, or agents.

Inputs

• Model (required) — Select the LM Studio embedding model to use for generating embeddings.
• Options — Configure additional embedding options: batch size (documents per request), whether to strip new lines from input text, request timeout in milliseconds, embedding dimensions (use -1 for model default), and encoding format ('float' for floating point arrays or 'base64' encoding).

Output shape

A single output named 'Embeddings' containing an array of embedding vectors corresponding to the input documents, produced in the selected encoding format.

Embeddings are returned in batches if input exceeds batch size. Each embedding corresponds to one input document. The node outputs through the 'ai_embedding' output channel. Connection to other AI nodes that consume embeddings is recommended for downstream use.

Examples

Example 1: Embedding multiple documents

Set Model to a valid LM Studio model. Set Batch Size to 100 to process inputs in chunks of 100. Provide an array of text documents as input to generate embeddings for each document. Strip New Lines true to sanitize input.