Nomic Embed Text v1.5

Nomic Embed Text v1.5 is an English text embedding model for developers building search and RAG applications. It represents text numerically so applications can find passages with similar meanings.

• Adjustable vector size: Smaller embeddings reduce storage needs, with a tradeoff in retrieval quality.
• Document retrieval: Separate task modes represent indexed documents and search queries according to their roles.
• Text organization: Embeddings support topic clustering, semantic duplicate detection, and inputs for classification models.
• Long passages: The model supports text sequences of up to 8,192 tokens.
• Text and image search: The companion nomic-embed-vision-v1.5 model shares its embedding space, so applications can compare text and image embeddings.

You can run the model on your own hardware through Sentence Transformers, Hugging Face Transformers, or Transformers.js, or serve it with Infinity in Docker. Local inference processes text on your hardware; the separate Nomic Embedding API sends requests to a hosted service.

The model is open source under Apache 2.0, and its training data is publicly released. Applications need to identify the embedding task when submitting text.

This listing is specifically nomic-embed-text-v1.5, whose model card identifies Apache 2.0. The Nomic Atlas Python client is a separate SDK and its release history is not the model’s version or maintenance history.