Generate embeddings

Convert text into vectors for subsequent storage, comparison, or retrieval in your application. This task covers vector generation; vector storage and index management are outside its scope.

Prerequisites

Prepare a personal access token or service account API Key with Genesis permissions. See Access credentials. Find and select a model for embeddings, and prepare the text to convert.

Call sequence

  1. Create an Embedding using the selected model and text. Preserve the original text order for batch input.

  2. Read the returned vectors and use their input indices to associate them with the original text.

  3. Pass the vectors and their corresponding text to the application’s next processing step.

API operations

Task

API operation

When to use it or what to check

Convert text into vectors

Create an Embedding

Process one text or a batch and obtain the corresponding vectors.

Confirm the result

Check that the results correspond to the input texts and that vector encoding and length meet your application’s needs.

Limitations and notes

Support for output encodings and custom vector dimensions depends on the selected model. Do not assume that every model supports the same options.

Last updated on