# Surya

Framework · https://godooo.ai/en/framework/surya

> Surya is an open-source AI framework, launched in 2024, made for developers.

An OCR toolkit for text detection, layout, reading order and tables in over 90 languages.

Main jobs: Document parsing, Text recognition (OCR). It is open source and free to use. You can use it on your own server, offline on your own machine and through an API.

## At a glance

- **Type:** [Framework](https://godooo.ai/en/browse#framework)
- **Launched:** 2024
- **Galaxy:** [Models & APIs](https://godooo.ai/en/galaxy/builders)
- **Open source:** Yes · Apache-2.0
- **Source code:** [github.com/datalab-to/surya](https://github.com/datalab-to/surya)
- **GitHub stars:** 21,415
- **How you run it:** Self-host · Runs offline · API
- **Made for:** Developers
- **Website:** [datalab.to](https://www.datalab.to)

## What it can do

- **[Document parsing](https://godooo.ai/en/capability/document-parsing)** — main job, _proposed by a machine, awaiting calibration_. “Layout analysis (table, image, header, etc.) with reading order - Table recognition (rows + columns)” (read on [github.com](https://github.com/datalab-to/surya), 2026-09-24)
- **[Text recognition (OCR)](https://godooo.ai/en/capability/ocr)** — main job, _proposed by a machine, awaiting calibration_. “Surya is a 650M param OCR model with these features: - Accuracy - scores 83.3% on olmOCR-bench (top under 3B params)” (read on [github.com](https://github.com/datalab-to/surya), 2026-09-24)

## Alternatives

- **[MinerU](https://godooo.ai/en/framework/mineru)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[PaddleOCR](https://godooo.ai/en/framework/paddleocr)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[Docling](https://godooo.ai/en/framework/docling)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[Marker](https://godooo.ai/en/framework/marker)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[MarkItDown](https://godooo.ai/en/framework/markitdown)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[Unstructured](https://godooo.ai/en/framework/unstructured)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[LlamaIndex](https://godooo.ai/en/framework/llamaindex)** (Open source, free): Also does: Document parsing, Text recognition (OCR)
- **[RAGFlow](https://godooo.ai/en/product/ragflow)** (Open source, free): Also does: Document parsing
- **[PDF Skill](https://godooo.ai/en/skill/pdf)**: Also does: Document parsing
- **[Firecrawl](https://godooo.ai/en/product/firecrawl)** (Open source, free): Also does: Document parsing
- **[Mistral](https://godooo.ai/en/model/mistral)** (Open source, free): Also does: Text recognition (OCR)
- **[Qwen](https://godooo.ai/en/model/qwen)** (Open source, free): Also does: Text recognition (OCR); no code needed

## Questions

### What is Surya?

Surya is an open-source AI framework, launched in 2024, made for developers. An OCR toolkit for text detection, layout, reading order and tables in over 90 languages.

### Is Surya free?

Yes. Surya is open source (Apache-2.0) and free to use.

### Is Surya open source?

Yes, under the Apache-2.0 license. The source code is at https://github.com/datalab-to/surya.

### Can Surya run locally?

Yes. Surya can run offline on your own machine. It can also be self-hosted on your own server.

### What can Surya do?

Main jobs: Document parsing, Text recognition (OCR).

### What are open-source alternatives to Surya?

Open-source ones: MinerU, PaddleOCR, Docling, Marker, MarkItDown, Unstructured, LlamaIndex and RAGFlow. Others: PDF Skill.

## Sources

- [github.com](https://github.com/datalab-to/surya) — checked 2026-09-24
- [datalab.to](https://www.datalab.to)

---

Last updated 2026-09-24 · sources checked 2026-09-24 · Markdown version of https://godooo.ai/en/framework/surya. Every page on this site has one: add .md to its address.