Starstruk Logo MarkStarstruk

v1.15.0

t8y2/dbxv1.15.0Aug 17, 2026by abimaelmartell

AI Summary

This release introduces selective OCR capabilities for scanned and mixed PDFs using PDFium rendering and PP-OCRv6 Small. It focuses on recovering text layers, optimizing OCR performance by reusing sessions, and exposing the pipeline across multiple languages with detailed metrics.

Key Highlights

  • Adds selective OCR for scanned pages and mixed PDFs using PDFium rendering and PP-OCRv6 Small through ONNX Runtime.
  • Recovers credible PDFium text layers before OCR, routes only affected pages, and adaptively fuses OCR with usable native text.
  • Reuses OCR sessions and preserves row boundaries for faster, cleaner extraction.
  • Exposes the OCR pipeline through Rust, pdf2md, Node.js, and Python, including page-level provenance, confidence, timings, and warnings.
  • Adds pinned runtime and model caching, offline operation, and cross-language smoke coverage.

New Features

  • Selective OCR using PDFium and PP-OCRv6 Small for scanned and mixed PDFs.
  • PDFium text layer recovery and adaptive fusion with native text.
  • OCR session reuse and row boundary preservation for efficiency.
  • Exposed OCR pipeline with detailed metrics (provenance, confidence, timings).
  • Pinned runtime, model caching, and offline operation support.

Full Release Notes

## Packages

All packages were built from source checkpoint [`06a9bab`](https://github.com/firecrawl/pdf-inspector/commit/06a9bab6b3309309503f2db17851389cee094a62).

- [Rust crate `pdf-inspector` 1.15.0](https://crates.io/crates/pdf-inspector/1.15.0) — `cargo add pdf-inspector@1.15.0`
- [Python `pdf-inspector` 1.15.0](https://pypi.org/project/pdf-inspector/1.15.0/) — `pip install pdf-inspector==1.15.0`
- [Node `@firecrawl/pdf-inspector` 1.15.0](https://www.npmjs.com/package/@firecrawl/pdf-inspector/v/1.15.0) — `npm install @firecrawl/pdf-inspector@1.15.0`
- [WebAssembly `@firecrawl/pdf-inspector-wasm` 1.15.0](https://www.npmjs.com/package/@firecrawl/pdf-inspector-wasm/v/1.15.0) — `npm install @firecrawl/pdf-inspector-wasm@1.15.0`

<details>
<summary>Node platform packages</summary>

- [Linux x64 glibc](https://www.npmjs.com/package/@firecrawl/pdf-inspector-linux-x64-gnu/v/1.15.0)
- [Linux x64 musl](https://www.npmjs.com/package/@firecrawl/pdf-inspector-linux-x64-musl/v/1.15.0)
- [Linux ARM64 glibc](https://www.npmjs.com/package/@firecrawl/pdf-inspector-linux-arm64-gnu/v/1.15.0)
- [Linux ARM64 musl](https://www.npmjs.com/package/@firecrawl/pdf-inspector-linux-arm64-musl/v/1.15.0)
- [macOS ARM64](https://www.npmjs.com/package/@firecrawl/pdf-inspector-darwin-arm64/v/1.15.0)
- [Windows x64 MSVC](https://www.npmjs.com/package/@firecrawl/pdf-inspector-win32-x64-msvc/v/1.15.0)

</details>

## Highlights

- Adds selective OCR for scanned pages and mixed PDFs using PDFium rendering and PP-OCRv6 Small through ONNX Runtime (#348, #357, #358).
- Recovers credible PDFium text layers before OCR, routes only affected pages, and adaptively fuses OCR with usable native text (#359, #393, #394).
- Reuses OCR sessions and preserves row boundaries for faster, cleaner extraction (#390, #391).
- Exposes the OCR pipeline through Rust, `pdf2md`, Node.js, and Python, including page-level provenance, confidence, timings, warnings, and hosted-parser recommendations (#360, #403, #405).
- Adds pinned runtime and model caching, offline operation, and cross-language smoke coverage (#404, #409).

## Other improvements

- Rejoins words hyphenated across line breaks (#388).
- Improves running-header, running-footer, and numbered-reference-note handling (#374, #392).

[Compare changes since 1.14.2](https://github.com/firecrawl/pdf-inspector/compare/v1.14.2...v1.15.0)