Venice AI

Use Venice AI for chat, image generation, embeddings, and more

Actions9

Embedding → Create

AI-generated

Overview

This node integrates with the Venice AI platform to create vector embeddings from text input. It sends a request to generate embeddings using a specified embedding model and returns the resulting embedding vectors. This is useful for scenarios such as semantic search, recommendation systems, or any application requiring numerical representations of text data.

Practical examples:

  • Converting product descriptions into embeddings for similarity-based product recommendations.
  • Generating embeddings of customer feedback for clustering or sentiment analysis.
  • Creating vector representations of documents to enable semantic search in a knowledge base.

Properties

Name Meaning
Model Embedding model ID to use for generating embeddings (e.g., "text-embedding-bge-m3").
Input Text string to generate embeddings for.
Encoding Format Format of the returned embedding vector: either "Float" (numeric array) or "Base64".

Output

The output JSON contains the response from the Venice AI embeddings API endpoint. Typically, this includes the generated embedding vector corresponding to the input text.

If the encoding format is:

  • Float: The embedding is returned as an array of floating-point numbers representing the vector.
  • Base64: The embedding is returned as a Base64-encoded string.

The exact structure depends on the Venice AI API response but generally includes fields like the embedding vector and metadata about the model used.

Dependencies

  • Requires an active Venice AI API key credential configured in n8n.
  • The node makes HTTP POST requests to the Venice AI /embeddings endpoint.
  • No additional environment variables are required beyond the API authentication setup.

Troubleshooting

  • Common issues:

    • Missing or invalid API key will cause authentication errors.
    • Providing empty or invalid text input may result in API errors or empty embeddings.
    • Selecting an unsupported embedding model ID can lead to failure responses.

  • Error messages:

    • Authentication failures: Check that the API key credential is correctly set up and valid.
    • "No binary data property found": Not applicable here since this operation does not handle binary data.
    • API request errors: Review the input parameters and ensure the model ID and input text conform to Venice AI API requirements.

Links and References

Discussion