FAQ

You can implement chunking strategies by specifying them in the chunking_strategy field when you create or update a data source. Current options include language-based, markdown-based, and HTML-based chunking.

To set up a model chain, use the create_chain endpoint. You’ll need to specify the types of models you want in the chain, the data source, and any chunking strategies you wish to apply.

While there’s no hard limit on the number of model chains or data sources, performance might vary based on server resources.

Custom HuggingFace models will be supported in future releases.

All data is encrypted during transit. We do not store your files on our servers unless explicitly instructed.

Log visibility is available between each model hand-off, aiding in debugging.