parse_pdf_stream
Parse a PDF file into JSON using configurable schemas. Output is delivered as a server-sent events stream for real-time processing.
Instructions
Parse one PDF as SSE stream through /api/v1/pdf/parse-stream.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| pages | No | ||
| use_ocr | No | ||
| pdf_path | Yes | ||
| schema_id | No | ||
| max_events | No | ||
| ocr_engine | No | surya | |
| schema_json | No | ||
| schema_name | No | ||
| response_mode | No | compact | |
| include_events | No | ||
| schema_template | No | ||
| event_preview_chars | No | ||
| max_estimated_tokens_override | No |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||