Nomic Atlas is a platform for interacting with unstructured datasets of text, image, video, audio, and embeddings at scale.
We've launched nomic-embed-text-v1.5 a text embedding model that supports variable output size!
Learn how to use Embeddings with Nomic Atlas.
- Learn how to effectively use the Atlas unstructured data interface.
- Learn about Embeddings and explore the Nomic API Reference.
- Get instant help on our Discord support channel.
- Read about our AI training policy and privacy guarantees.
Nomic Atlas enables anyone to find insights in and build with unstructured data. Unstructured data is anything that you normally would not store in a spreadsheet: large collections of text documents, galleries of images, audio files, videos and the training/evaluation datasets of AI models. Nomic Atlas uses AI and Embeddings to help you quickly understand, build with and share your unstructured datasets.
An embedding is a vector representation of an unstructured datapoint that enables computers to manipulate the data based on semantics and meaning. All data uploaded to Nomic Atlas has a corresponding embedding assigned to it. Nomic Embedding Models are used to assign embeddings to your uploaded data if you do not specify embeddings during data upload. Embeddings are key to powering the Unstructured Data Map and organizing your data by meaning.
Unstructured Data Map
The fastest way to understand and work with unstructured data is to look at it. Anytime you upload a datapoint, Nomic Atlas organizes it in an AI powered data interface called the map. The map groups together similar datapoints in your dataset spatially. You can collaborate on your dataset with others by sharing a browser link to the map and developers can access its data and operations programmatically.