API Documentation
Integrate data extraction into your applications with our REST API.
Quick Start
Upload a file and extract structured data in three API calls:
1
POST /api/upload/presigned→ Get upload URL2
PUT {uploadUrl}→ Upload file to S33
POST /api/upload/complete→ Confirm upload4
POST /api/process/{jobId}→ Start AI extraction (SSE stream)Supported File Types
PDFPNGJPG/JPEGTXTCSVJSONDOCX
auto
AI determines best approach
text
Extract clean text
structure
Convert to JSON
entity
Extract named entities
table
Extract tabular data
Endpoints
Large File Upload (Multipart, >100MB)
For files larger than 100MB, use the multipart upload flow:
POST /api/upload/multipart/initiate → Get uploadId and cloud_storage_path
POST /api/upload/multipart/part → Get presigned URL for each 10MB chunk
POST /api/upload/multipart/complete → Complete upload with ETags from each part
POST /api/process/{jobId} → Start extraction