Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.mindsdb.com/llms.txt

Use this file to discover all available pages before exploring further.

This documentation describes the integration of MindsDB with LangChain, a framework for developing applications powered by language models. The integration allows for the deployment of LangChain models within MindsDB, providing the models with access to data from various data sources.

​
Prerequisites

Before proceeding, ensure the following prerequisites are met:
  1. Install MindsDB locally via Docker or Docker Desktop.
  2. To use LangChain within MindsDB, install the required dependencies following this instruction.

​
Setup

Create an AI/ML engine from the LangChain Embedding handler. 
CREATE ML_ENGINE embedding
FROM langchain_embedding;
Create a model using embedding as an engine and providing your OpenAI API key. 
CREATE MODEL embedding_model
PREDICT embeddings
USING
      engine = "embedding",
      class = "embedding_model_provider",
      model = "embedding_model_name",
      openai_api_key = "sk-xxx",
      input_columns = ["column1", "column2", ...];
The following are the required and optional parameters:
  • engine is a required parameter. It defines the AI engine, as created with the CREATE ML_ENGINE statement, to be used.
  • class is a required parameter. It defines the model provider, such as "OpenAI" or "HuggingFace".
  • model is a required parameter. It defined the embedding model, such as text-embedding-3-small.
  • openai_api_key is a required parameter when using OpenAI as a provider.
  • input_columns is an optional parameter. It defines the column(s) to be processed by the embedding model.

​
Usage

The following usage examples utilize embedding to create a model with the CREATE MODEL statement.
  • Using the OpenAI models:
CREATE MODEL openai_embedding_model
PREDICT embeddings
USING
      engine = "embedding",
      class = "OpenAIEmbeddings",
      model = "text-embedding-3-small",
      openai_api_key = "sk-xxx",
      input_columns = ["content"];
  • Using the HuggingFace models:
CREATE MODEL hf_embedding_model
PREDICT embedding
USING
      engine = 'embedding',
      class = 'HuggingFaceEmbeddings',
      model = "sentence-transformers/all-mpnet-base-v2";
Ensure that the model has been created successfully before using it. To do that, use the DESCRIBE command and look at the status column. 
DESCRIBE openai_embedding_model;
Next StepsGo to the Use Cases section to see more examples.