MinerU 2.0 document parsing utility class Supports parsing PDF and image documents, converting the content into structured data and gene

Batch document parser with parallel processing capabilities Supports processing multiple documents concurrently with progress tracking a

Docling document parsing utility class. Specialized in parsing Office documents and HTML files, converting the content into structured d

Extracted general document content.

Container for tables and figures extracted from original document. Attributes: tables: List of extracted tables figures:

A single section/slide in the output.

Information about an extracted figure.

Generated image result.

Extracted paper content.

Information about an extracted table.

Result of batch processing operation

Planned content structure for generation.

Configuration for context extraction

Universal context extractor supporting multiple content source formats

Figure reference for a section.

User configuration for generation. Attributes: output_type: Type of output (poster or slides) poster_density: Content de

Complete input for generation. Attributes: config: User generation config content: PaperContent or GeneralContent from s

Configuration for Markdown to PDF conversion

Output type for generation.

Content density level for poster.

Processed custom style from LLM.

Configuration class for RAGAnything with environment variable support

Page count level for slides.

Predefined style types.

Table reference for a section.

Plans content structure using multimodal LLM.

Enhanced Markdown to PDF converter with multiple backends Supports multiple conversion methods: - WeasyPrint (recommended for HTML/CSS s

Summary with separated text and original elements. Attributes: summary_text: Text-based summary content (for text LLM) o

Processor specialized for equation content

Generic processor for other types of modal content

Generate poster/slides images from ContentPlan.

Processor specialized for image content

catch mineru error

Multimodal Document Processing Pipeline - Complete document parsing and insertion pipeline

RAG client for document indexing and querying. Example: async with RAGClient() as rag: await rag.index("document.pdf

Complete RAG configuration for Paper2Slides. All defaults are aligned with RAGAnythingConfig. Example: config = RAGConfig()

Storage paths for RAG index and parsed outputs.

Processor specialized for table content

LLM and Embedding API settings. API key must be provided via environment variable or explicitly. Base URL is optional (defaults to OpenAI

Base class for modal processors

Batch processing settings. Aligned with RAGAnythingConfig defaults.

BatchMixin class containing batch processing functionality for RAGAnything

Context extraction settings. Aligned with RAGAnythingConfig defaults.

Document processing status

Base class for document parsing utilities. Defines common functionality and constants for parsing different document types.

Paper parser settings.

ProcessorMixin class containing document processing functionality for RAGAnything

QueryMixin class containing query functionality for RAGAnything

Single RAG query result from batch_query.

RAG query results organized by section.

Document source type.

Configuration for summary processing.