AI Document Intelligence API
The Enterprise PII Detection & Redaction API helps developers and businesses securely identify, extract, mask, or redact sensitive data from documents and images in real time. Designed for enterprise-grade workflows, the API supports intelligent document analysis with high accuracy and scalable processing. Key Features AI-powered Personally Identifiable Information (PII) detection Automatic…
AI Document Intelligence API endpoints
| Method | Endpoint | Description |
|---|---|---|
| POST |
extractDocumentBatch /extract/batch |
Submit up to 20 documents for asynchronous extraction. Returns a `job_id` immediately; poll `GET /jobs/{job_id}` until status is `completed`. |
| POST |
extractDocument /extract |
Synchronously extract text from a single document (PDF, DOCX, PPTX, XLSX) using Mistral Document AI. For multiple files use `/extract/batch`. |
| POST |
transcribeMedia /transcribe |
Transcribe a single audio or video file (mp3, mp4, mpeg, mpga, m4a, wav, webm) via Whisper. Files are capped at **25 MB**. Returns a `job_id`; poll `GET /jobs/{job_id}` for the… |
| POST |
ocrImageBatch /ocr/batch |
Submit up to 20 images for asynchronous OCR. Returns a `job_id` to poll with `GET /jobs/{job_id}`. |
| POST |
ocrImage /ocr |
Run OCR on a single image (PNG, JPG, JPEG, TIFF, WEBP). |
| POST |
transcribeMediaBatch /transcribe/batch |
Submit multiple base64-encoded audio/video sources at once. Each source is sent to Whisper in parallel; per-file failures (bad base64, oversize, unsupported type) do not fail the… |
| GET |
getUsage /usage |
Return today's and this month's request and token usage for the calling organization, along with its tier limits. |
| GET |
getJobStatus /jobs/{job_id} |
Poll an async job (extract, OCR, or transcription). Status transitions `pending` → `processing` → `completed` or `failed`. Once completed, `results` contains one entry per… |
| GET |
getMyUsage /me/usage |
Same payload shape as `/usage`, but authenticated via the portal JWT instead of an API key. Useful for in-portal dashboards. |
| GET |
listPatterns /patterns |
List all PII regex patterns — built-in plus any user-defined ones. Optionally filter by business unit. |
| GET |
listEntities /entities |
List every PII entity type the redactor can detect. Includes built-in entities and any active custom patterns registered for the calling organization. |