oxidize-pdf
Pure Rust PDF library for structure-aware RAG, parsing, and generation.
Process, parse, and generate PDFs with structure-aware chunking and zero C dependencies.
oxidizePdf is a pure Rust PDF toolkit tailored for document processing and AI ingestion pipelines. It gives you structure-aware chunks with metadata like bounding boxes and heading contexts without needing native C bindings.
What it is
The library parses PDFs, extracts text, handles digital signatures, validates PDF/A levels, and partitions documents into typed elements. It provides specialized RAG pipelines that compute token estimates and inject deterministic context prefixes for vector store ingestion.
Who it's for
It suits Rust developers and engineers building high-performance document ingestion pipelines or AI search systems. It does not suit teams looking for a page rasterizer or out-of-the-box UI rendering tools.
Key features
Structure-Aware Chunking
Extracts chunks with page numbers, bounding boxes, element types, and token estimates.
Contextual Retrieval Support
Prepends deterministic document and section context prefixes without requiring an ML model.
Pure Rust Architecture
Handles PDF workflows and certificate verification entirely in Rust without C bindings.
Document Partitioning
Breaks pages down into typed elements and relationship graphs for fine-grained extraction.
Up and running in 3 steps
Our honest verdict
What's great
- No native C dependencies or external crypto libraries required for core operations.
- Preserves heading context and bounding boxes directly in RAG chunks.
- Performs structure-aware parsing and invoice extraction across multiple languages.
- Integrates cleanly into Rust projects via straightforward Cargo dependencies.
Where it falls short
- Requires writing and maintaining Rust code to integrate into your data pipeline.
- Does not provide native page rasterization, requiring external renderers for bitmap outputs.
- Certificate-based recipient encryption is unsupported, throwing explicit errors.
Pricing & license
Free under its MIT licence, with the only cost being the hardware and time required to run it yourself.
Open remote AI-training roles
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
Use it, or have us set it up
FAQ
Does oxidize-pdf require C dependencies?
No, it is written entirely in pure Rust for both PDF processing and certificate verification.
Can I rasterize pages into images with this library?
No, page rasterization is outside the scope of the library; you must use an external renderer like Pdfium.
How does contextual retrieval work without ML?
It appends a deterministic text prefix based on document metadata and heading breadcrumbs to each chunk.
