nomic-embed-text embeddings locally with Ollama

nomic-embed-text for a 100% local English RAG with Ollama: 768 dimensions, long texts, search_query: / search_document: prefixes. Apache 2.0.

Notes

Available in Ollama: ollama pull nomic-embed-text, then call /api/embed.
The search_query: and search_document: prefixes are recommended by the authors.
Trained mostly on English, which makes it a good default for English documents.
Published by Nomic AI under the Apache 2.0 license.

Estimated cost

Input tokens per call : ≈ 236

Input cost only, excluding the model's answer. Providers' public standard prices (Mistral AI, DeepSeek, OpenAI, Anthropic) converted from USD to euros at the ECB rate of September 29, 2026. Estimate: 1 token ≈ 3.6 characters.

Identifier

contextetech--nomic-embed-text-ollama-en

Type

Embeddings

File

nomic-embed-text-ollama-en.embedding.json

Format

JSON (application/json)

Size

619 bytes

Encoding

UTF-8

Estimated tokens

≈ 172

Language

English

License

MIT

Commercial use

Allowed

Personal data

None

Version

1.0

Published on

October 2, 2026 at 9:54 PM

Updated on

October 2, 2026 at 9:54 PM

Storage

in database

Uses

0

Likes

0

nomic-embed-text-ollama-en.embedding.json

{
  "format": "contextetech/embedding/v1",
  "id": "contextetech/nomic-embed-text-ollama-en",
  "description": "nomic-embed-text for a 100% local English RAG with Ollama: 768 dimensions, long texts, search_query: / search_document: prefixes. Apache 2.0.",
  "tags": [
    "ollama",
    "rag",
    "local"
  ],
  "license": "MIT",
  "language": "en",
  "model": "nomic-ai/nomic-embed-text-v1.5",
  "dimensions": 768,
  "maxTokens": 8192,
  "distance": "cosine",
  "languages": [
    "en"
  ],
  "prefixes": {
    "query": "search_query: ",
    "passage": "search_document: "
  },
  "chunkSize": 500,
  "benchmarks": []
}

Embed (Python)

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("nomic-ai/nomic-embed-text-v1.5")
q = model.encode(["search_query: your input here"], normalize_embeddings=True)
docs = model.encode(["search_document: …"], normalize_embeddings=True)
print(q.shape)  # (1, 768)
print(q @ docs.T)  # similarité