# ScrapingBee YouTube Search API

> Search YouTube and receive raw renderer data for matching results.

## Endpoint and authentication

- Endpoint: `https://app.scrapingbee.com/api/v1/youtube/search`
- Required input: `search`.
- Authentication: `Authorization: Bearer YOUR_API_KEY`.
- The `api_key` query parameter remains supported for backward compatibility but is deprecated for new integrations.
- Cost: 5 credits per successful request.

## Parameters and response

- URL-encode `search`, especially special characters.
- `upload_date`: `today`, `last_hour`, `this_week`, `this_month`, or `this_year`.
- `type`: `video`, `channel`, `playlist`, or `movie`.
- `duration`: `<4`, `4-20`, or `>20`.
- `sort_by`: `relevance` (default), `rating`, `view_count`, or `upload_date`.
- Boolean filters: `hd`, `4k`, `subtitles`, `creative_commons`, `live`, `360`, `3d`, `hdr`, `location`, `purchased`, and `vr180`.
- `tag` is an optional response-header label and does not change scraping behavior.
- The source describes the result as parsed but its documented payload is raw YouTube renderer data in `results`; it echoes the `search` query. Treat renderer structures as YouTube-controlled and potentially variable, rather than relying on a fixed normalized schema.
- The source does not document page, country, or language controls for this endpoint.
- Failed searches are retried by ScrapingBee for up to 30 seconds. Account for that server-side retry window in client timeouts.