> ## Documentation Index
> Fetch the complete documentation index at: https://pythonaisdk.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Vector search

> Embed a corpus and find the most similar sentence via cosine similarity.

<Steps>
  <Step title="Embed your corpus">
    <CodeGroup>
      ```python search.py theme={null}
      from ai_sdk import openai, embed_many, cosine_similarity

      sentences = [
      "The cat sat on the mat.",
      "A dog was lying on the rug.",
      "Paris is the capital of France.",
      "Cats love sunny spots.",
      ]
      model = openai.embedding("text-embedding-3-small")
      res = embed_many(model=model, values=sentences)

      ```
    </CodeGroup>
  </Step>

  <Step title="Search for a query">
    ```python theme={null}
    query = "Dogs enjoy napping on carpets."
    q_emb = embed_many(model=model, values=[query]).embeddings[0]

    scores = [cosine_similarity(q_emb, emb) for emb in res.embeddings]
    most_similar_idx = max(range(len(scores)), key=scores.__getitem__)
    print("Most similar: ", sentences[most_similar_idx])
    ```

    Expected result:

    ```text theme={null}
    Most similar:  A dog was lying on the rug.
    ```
  </Step>
</Steps>

<Note>
  `cosine_similarity` is dependency-free. Replace it with NumPy or faiss in production for better
  performance.
</Note>