URL
Index content from a specified URL with optional metadata and processing settings.
The Index by URL endpoint allows you to index content from a specified URL. This is useful for processing and indexing remote files or web pages directly.
Request
The URL of the content to be indexed
The ID of the collection to index the content into
Whether to save the indexed content (default: true)
Settings for processing video content
Video Settings Object
The interval (in seconds) at which to split the video for processing
Settings for reading the text in each video interval
Settings for embedding each video interval
Settings for transcribing each video interval
Settings for describing the content in each video interval
Settings for generating structured JSON output for each video interval
Each processing method (read, embed, transcribe, describe) can have a model_id specified. The available models can be found in the Models section.
Additionally, describe and json_output methods can include a prompt to guide the output generation.
Response
A message indicating the status of the indexing process
A unique identifier for the indexing task
The indexing process is asynchronous. You can use the returned task_id to check the status of the indexing task if needed.
Was this page helpful?