Reducto API Overview
Welcome to Reducto! This guide provides an overview of our API endpoints and answers frequently asked questions. For detailed information on individual endpoints, please refer to our endpoint-specific recipes.
Core Processing Endpoints
We have two main processing endpoints at Reducto depending on your use case. Both endpoints support documents, slides, spreadsheets, and images.
1. /parse
Purpose: Parse all content in a document and handle post-processing steps like chunking.
Key Features:
- Only requires a URL to the document
- Ideal for use cases such as RAG, summarization, or any pipeline using full document reasoning
2. /extract
Purpose: Perform structured JSON extraction to pull out specific fields according to your schema.
Key Features:
- Uses /parse alongside LLM calls to map document information into your schema
- Ideal for process automation or extracting data across document corpora (e.g., pulling purchases from invoices)
Supporting Endpoints
- /upload: Upload files for use with Reducto's API without external hosting services like S3.
- /job/{job_id}: Poll for the status of a given asynchronous job.
- /configure_webhook: Configure your webhook via SVIX.
Updated about 1 month ago
What’s Next
Our processing endpoints offer various configuration options to optimize performance for your specific use case. You can learn more at: