Overview
The HyperFlow LLM PDF Transformer provides intelligent PDF-to-Markdown conversion using large language models with two distinct processing modes optimized for different use cases. The service combines computer vision, traditional PDF processing, and LLM capabilities to deliver high-quality document transformation.
Processing Modes:
- Image Rendering Mode: Traditional page-by-page processing that renders PDF pages as images for LLM analysis. Best for scanned documents, complex layouts, or when embedded image extraction is required.
- Direct Upload Mode: Streamlined processing that sends PDF files directly to LLMs that have native PDF support. Offers up to 3-5x faster processing and batch capabilities, ideal for text-heavy documents and modern LLM workflows.
Both modes support custom transformations through instruction inputs, enabling specialized data extraction, JSON/HTML output generation, and sophisticated handling of complex document structures beyond standard Markdown conversion.
Service Information
| Property |
Value |
| Service Name |
HyperFlow LLM PDF transformer |
| Status |
Enabled |
| Compatible Nodes |
Transform Content |
When to Use This Service
The HyperFlow LLM PDF Transformer excels at:
General Use Cases:
- Standard PDF-to-Markdown conversion with superior accuracy compared to traditional OCR
- Complex document layouts with mixed content (text, images, tables, equations)
- Mathematical content with automatic LaTeX/Markdown formatting
- Multi-page contextual analysis maintaining understanding across document sections
- Advanced custom transformations through instruction inputs
Image Rendering Mode Specific:
- Scanned documents requiring intelligent OCR with layout understanding
- Embedded image extraction with automatic content store integration
- Page range processing for selective document analysis