---
title: "Search models, collections, and docs (beta)"
method: GET
path: "/search"
---

# Search models, collections, and docs (beta)

`GET /search`

Search for public models, collections, and docs using a text query.

For models, the response includes all model data, plus a new `metadata` object with the following fields:

- `generated_description`: A longer and more detailed AI-generated description of the model
- `tags`: An array of tags for the model
- `score`: A score for the model's relevance to the search query

Example cURL request:

```console
curl -s \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  "https://api.replicate.com/v1/search?query=nano+banana"
```

Note: This search API is currently in beta and may change in future versions.

## Query parameters

- `query` string, required
- `limit` integer

## Response `200`

Success

- SchemasSearchResponse
  - `collections` SchemasSearchCollectionResponse[], required — Array of collections that match the search query
    - `description` string, required — A description of the collection
    - `models` string[], nullable — Array of model names in the collection
    - `name` string, required — The name of the collection
    - `slug` string, required — The slug of the collection (lowercase with dashes)
  - `models` object[], required — Array of models that match the search query, each containing model data and extra metadata
    - `metadata` SchemasSearchModelMetadataResponse, required
      - `generated_description` string, nullable — AI-generated detailed description of the model
      - `score` number, nullable — Search relevance score
      - `tags` string[] — Array of descriptive tags for the model
    - `model` SchemasModelResponse, required
      - `cover_image_url` string, uri, nullable — A URL for the model's cover image
      - `default_example` object, nullable — The model's default example prediction
      - `description` string, nullable — A description of the model
      - `github_url` string, uri, nullable — A URL for the model's source code on GitHub
      - `is_official` boolean — Boolean indicating whether the model is officially maintained by Replicate. Official models are always on, have stable API interfaces, and predictable pricing.
      - `latest_version` object, nullable — The model's latest version
      - `license_url` string, uri, nullable — A URL for the model's license
      - `name` string — The name of the model
      - `owner` string — The name of the user or organization that owns the model
      - `paper_url` string, uri, nullable — A URL for the model's paper
      - `run_count` integer — The number of times the model has been run
      - `url` string, uri — The URL of the model on Replicate
      - `visibility` 'public' | 'private' — Whether the model is public or private
  - `pages` SchemasPageResponse[], required — Array of Replicate documentation pages that match the search query
    - `href` string, required — URL path to the page
    - `name` string, required — Title of the page
  - `query` string, required — The search term that was evaluated

## Other responses

- `400` — Bad Request - Missing or invalid query parameter
- `500` — Internal Server Error - Search service error

---

[API](https://skmtc.net/replicate/apis/replicate-http-api.md) · [All operations](https://skmtc.net/replicate/apis/replicate-http-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/replicate/replicate-http-api/versions/7a537f433b0b/schema)