LaTeX to JSON Converter
Convert LaTeX papers to structured JSON with full document hierarchy and metadata.
Free for LaTeX uploads. Plus subscription required for JSON export.
Free account required. See pricing for high-volume use.
Built for your workflow
LaTeX to JSON conversion that actually works for academic papers.
LLM Pipelines
Build RAG systems with structured paper data - query specific sections, equations, or citations
Paper Analysis
Extract all equations, count citations, analyze document structure programmatically
Custom Rendering
Build your own paper viewer or transform papers into any format you need
Data Pipelines
Process thousands of papers with consistent, machine-readable structure
Semantic parsing
We understand LaTeX structure, not just text. That's why our output preserves what matters.
Semantic Parsing
Understands LaTeX structure - sections, equations, theorems, figures, tables as semantic elements
Cross-Reference Resolution
Automatically resolves \ref, \cite, and other cross-references to readable formats
Macro Expansion
Expands custom macros and commands so the output is self-contained
Bibliography Support
Includes formatted references with proper numbering and citation links
Hierarchical Structure
Full document tree with sections, subsections, and nested content blocks
Type-Safe Schema
Every element has a type field - section, equation, figure, table, citation, etc.
See what you get
Real output from converting the “Attention Is All You Need” paper.
{
"by": "sciencestack.ai",
"title": "Attention Is All You Need",
"abstract": "The dominant sequence transduction models...",
"document": [
{
"type": "section",
"title": "Introduction",
"content": [
"Recurrent neural networks, long short-term memory \\cite{bib:1}...",
{
"type": "figure",
"src": "assets/transformer.png",
"caption": "The Transformer model architecture."
}
]
},
{
"type": "section",
"title": "Attention",
"content": [
"An attention function maps a query and key-value pairs...",
{
"type": "equation",
"content": "\\text{Attention}(Q,K,V) = \\text{softmax}(\\frac{QK^T}{\\sqrt{d_k}})V"
}
]
}
],
"bibliography": [...]
}Equations, cross-references, and structure — all preserved.
How it works
Upload
Upload your .tex file or .zip archive with all source files
Process
We parse the LaTeX semantically — understanding sections, equations, and references
Download
Structured JSON with document hierarchy, equations as LaTeX strings, and rich metadata
Simple pricing
For LaTeX conversions
- Full LaTeX parsing
- Equation preservation
- Cross-reference resolution
- Bibliography included
- Structured document tree
JSON export requires a Plus subscription.