Atlas Datasets

See the frontier of multimodal model training by exploring this 11M document subset of OBELICS: an open collection of interleaved image-text web documents, containing 141M English documents, 115B text tokens, and 353M images. Created in a collaboration between Nomic and Huggingface.

Explore 5.4 million of the most popular English-language tweets from the last 13 years, arranged thematically with date information.

Explore KREA's stable diffusion database to learn how people are using generative AI to create new images.

Explore an interactive map by HacDC's hyperdemocracy group, featuring 700,000 text snippets from ten years of U.S. Congressional legislation. The map includes detailed metadata such as political party, state, and NOMINATE scores, helping you discover the contents of proposed laws.

Explore the changing landscape of machine learning research with this text map of all accepted submissions to NeurIPS between 1987 to 2023.

An explorable map of 14,732 submissions to the ICLR conference, from 2018-2023.

Google's Vertex AI group used Nomic to visualize their embeddings of 8 million stack overflow posts. Explore where people get stuck while coding, and what answers they got from 'hello worlds' in every language to assembly language and Terraform configurations.