v2.63.0

docling-project/doclingv2.63.0Nov 20, 2025by docling-ops[bot]

AI Summary

The update adds save and load functionality for conversion results and fixes issues with threaded pipelines, document timeouts, and GPU availability for RapidOCR. It also updates documentation for various integrations.

Key Highlights

  • Add save and load for conversion result
  • Respect document_timeout in new threaded StandardPdfPipeline
  • Enable GPU for RapidOCR when available
  • Remove py3.14 requirement for default rapidocr
  • Add Hector as compatible AI agent platform integration

New Features

  • Save and load for conversion result

Full Release Notes

### Feature

* Add save and load for conversion result ([#2648](https://github.com/docling-project/docling/issues/2648)) ([`b559813`](https://github.com/docling-project/docling/commit/b559813b9becf7950bc539c1334e55ef17bed2ad))

### Fix

* Respect document_timeout in new threaded StandardPdfPipeline ([#2653](https://github.com/docling-project/docling/issues/2653)) ([`2087c6b`](https://github.com/docling-project/docling/commit/2087c6bf9f65f279dd2ff0631768996aecd640fe))
* In DocumentConverter.convert_string() make nullable name parameter optional ([#2660](https://github.com/docling-project/docling/issues/2660)) ([`6fb9a5f`](https://github.com/docling-project/docling/commit/6fb9a5f98aa33f0e37d720487e71467cd9aae846))
* Enable GPU for RapidOCR when available ([#2659](https://github.com/docling-project/docling/issues/2659)) ([`463a3fd`](https://github.com/docling-project/docling/commit/463a3fd4743708ccebe32fa1ddb462a6899277b9))
* Remove py3.14 requirement for default rapidocr ([#2639](https://github.com/docling-project/docling/issues/2639)) ([`da4c2e9`](https://github.com/docling-project/docling/commit/da4c2e9dbe5ea363d29af188e24885f1ef3efba9))

### Documentation

* Add Hector as compatible AI agent platform integration ([#2662](https://github.com/docling-project/docling/issues/2662)) ([`ce5a099`](https://github.com/docling-project/docling/commit/ce5a099dfde50b5416f7a9676914c938740e0ada))
* Added documentation to use SuryaOCR via plugin docling-surya ([#2533](https://github.com/docling-project/docling/issues/2533)) ([`b216ad8`](https://github.com/docling-project/docling/commit/b216ad848dfe6d8cdfdbb90d10000773d8563633))
* Fix broken homepage links ([#2651](https://github.com/docling-project/docling/issues/2651)) ([`03e7c7d`](https://github.com/docling-project/docling/commit/03e7c7d924f4973c679ef988bc861741e02d804a))
* **examples:** Processing parquet file of images ([#2641](https://github.com/docling-project/docling/issues/2641)) ([`8af228f`](https://github.com/docling-project/docling/commit/8af228f1e2e5f12815f19beb979e7b1253c15115))
* Move Installation and Quickstart (Usage) under Getting started ([#2644](https://github.com/docling-project/docling/issues/2644)) ([`d549445`](https://github.com/docling-project/docling/commit/d549445e78fc8839c510a52e6c72e2e711ba153e))
* Add redirection from getting started page ([#2640](https://github.com/docling-project/docling/issues/2640)) ([`ac9fc58`](https://github.com/docling-project/docling/commit/ac9fc585bbb515c3e00b042eeb979800ae548660))
* **examples:** Remove deprecation warnings with export_to_dataframe ([#2638](https://github.com/docling-project/docling/issues/2638)) ([`f552862`](https://github.com/docling-project/docling/commit/f5528623a77d9a3b0d821035f66baefd49da8dd9))