indra vs docling
CPU-focused OCR and broad document conversion compared
indra and docling turn documents into structured, usable content. indra focuses on fast CPU OCR for PDFs and images, while docling handles a wider range of formats and document-processing workflows.
indra: Fast CPU OCR with Datalab’s Indra Model
Indra OCR uses Datalab’s compact model with ONNX Runtime to turn PDFs and images into markdown, text, or structured JSON on CPU. It adds line boxes, layout labels, and reading order for fast document parsing without a GPU.
docling: Convert Documents into Structured Content
Docling parses documents such as PDFs, Word files, spreadsheets, and images into structured content for applications and generative AI workflows. It offers local processing, a Python library, and a command-line interface.
| indra | docling | |
|---|---|---|
| Language | Python | Python |
| License | Apache-2.0 | MIT |
| Stars | 421 | 68.7k |
| Forks | 33 | 5k |
| Last analyzed | Oct 16, 2026 | Oct 12, 2026 |
Key differences
- indra is focused on PDFs and images; docling also parses office, web, audio, video, and other document types.
- indra uses a compact ONNX Runtime model without requiring PyTorch or a GPU; docling supports OCR and visual-language-model pipelines, with local processing available.
- indra outputs Markdown, plain text, or JSON and includes line boxes, confidences, layout labels, and reading order; docling offers formats including Markdown, HTML, WebVTT, DocTags, and lossless JSON.
- indra’s code is Apache-2.0, while its model weights use a modified OpenRAIL-M license with additional terms; docling’s code is MIT-licensed, and model packages can have separate licenses.
- indra supports Linux and macOS and is aimed at clean printed documents; docling targets broader document-processing and generative AI workflows.
- The supplied health data describes indra as very new, with contributions from one person; docling has regular commits and releases, with contributions not concentrated in one person.
Choose indra if you…
- need CPU-focused OCR for clean printed PDFs and images without a GPU.
- want line-level coordinates, confidences, layout labels, and reading order in JSON output.
- process digital PDFs alongside scanned pages and want a local viewer to inspect extracted results.
Choose docling if you…
- need to process a broad mix of PDFs, office files, web formats, images, audio, or video.
- build document or generative AI workflows using integrations, a CLI, or a Python API.
- want local or air-gapped processing and exports such as Markdown, HTML, or lossless JSON.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.