Text Embeddings

Nomic Text Embeddings allow you to semantically encode text for computers to manipulate. They are useful for understanding large unstructured datasets, semantic search and building retrieval-augmented LLM apps.

Explore all Nomic Text Embedding models and use them in Python.

Get your Nomic Atlas API key at atlas.nomic.ai.

Create Text Embeddings

POST

https://api-atlas.nomic.ai/v1/embedding/text

Creates embeddings from a batch of text documents. Optionally, specify the embedding task_type to specialize the embeddings to a certain task. Notably, RAG workflows should use search_query for queries and search_document for documents.

Request Body

model

required

string

The Nomic Text Embedding model to use.

texts

required

list

A list of text documents to embed

task_type

optional

string

The task your embeddings should be specialized for: search_query, search_document, clustering, classification. Defaults to search_document.

dimensionality

optional

integer

The output size of the embedding model. Applicable only to models that support variable dimensionality and defaults to the models largest embedding size.

Returns

The list of text embeddings and the total tokens utilized by the request.

Request Example

Copy

curl https://api-atlas.nomic.ai/v1/embedding/text \
    -H "Authorization: Bearer $NOMIC_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{ "model": "nomic-embed-text-v1", "texts": ["The quick brown fox..."]}'

Response Example

Copy

{
  "embeddings": [[1.320, 1.232, ...]], 
  "usage": {"total_tokens": 7}
}