Nomic AI develops the Nomic Embed text embedding model.

Nomic Embed (nomic-embed-text-v1.5) is an open, 8192-token-context text embedding model that outperforms OpenAI's ada-002 on retrieval benchmarks.

Weights are on Hugging Face under nomic-ai, with training code on GitHub at nomic-ai/contrastors.

Released in February 2024, with the Matryoshka-dimension v1.5 update following shortly after.

Built to provide a fully open, reproducible embedding model with auditable training data for retrieval and RAG applications.

Used via Sentence-Transformers, Hugging Face Transformers, or the Nomic Atlas API.

Apache 2.0 license, fully open weights and training data, no credentials required, self-hosted or via Nomic's hosted API; maintained by Nomic AI.