Using a Self-Hosted PDF OCR API with PaddleOCR

paddleocr-pdf-api is an open-source Docker image that wraps PaddleOCR's vision-language model (PaddleOCR-VL-1.5, 0.9B parameters) into a self-hosted REST API running on a local GPU (~8.5 GB VRAM). It processes PDFs page-by-page through a job queue, streaming results as Markdown without requiring cloud services. The API supports job submission, progress polling, per-page retrieval, cancellation, and optional API key authentication.