Sparse Reference Sliding Window Attention (SR-SWA): Compresses text-visual attention interactions from ultra-long sequences down to a fixed-size set of working references (e.g., 512 tokens).
Alternatives to Unlimited-OCR-2
120 open-weight models that do the same job as Unlimited-OCR-2 (3.3B OCR), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.
PP-OCRv6 is a lightweight OCR system that combines architectural innovation with data-centric optimization. It redesigns the backbone, detection neck, and recognition neck around a unified MetaFormer-style building block with structural reparameterization.
?BApache 2.012.9K 18Jun 2026- Same category
- Commercial use OK
PP-OCRv5_server_rec is one of the PP-OCRv5_rec that are the latest generation text line recognition models developed by PaddleOCR team.
?BApache 2.0159.1K 32Jun 2025- Same category
- Commercial use OK
Merged variant for extra robustness (advanced). This model combines the base checkpoint with RLVR-trained weights using task arithmetic merging, providing improved consistency across challenging document categories.
1B16K ctx Apache 2.0392 14Jan 2026- Same category
- Smaller (1B vs 3.3B)
- Commercial use OK
This Hub repository contains a HuggingFace's transformers implementation of Florence-2 model from Microsoft.
777M4K ctx MIT353.9K 1.9KJun 2024- Same category
- Smaller (777M vs 3.3B)
- Commercial use OK
PP-OCRv5_mobile_det is one of the PP-OCRv5_det series, the latest generation of text detection models developed by the PaddleOCR team.
?BApache 2.091.2K 33Jun 2025- Same category
- Commercial use OK
PP-OCRv5_mobile_rec is one of the PP-OCRv5_rec that are the latest generation text line recognition models developed by PaddleOCR team.
?BApache 2.052.2K 17Jun 2025- Same category
- Commercial use OK
Document parsing for English and 22 Indian languages, printed and handwritten. A page image in; reading-ordered Markdown out, with math as LaTeX and tables as HTML or Markdown, plus per-block JSON.
?BCustom licence1.3K 79Aug 2026- Same category
General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
716M32K ctx Apache 2.0124.7K 1.6KSept 2024- Same category
- Smaller (716M vs 3.3B)
- Commercial use OK
PP-OCRv6 is a lightweight OCR system that combines architectural innovation with data-centric optimization. It redesigns the backbone, detection neck, and recognition neck around a unified MetaFormer-style building block with structural reparameterization.
?BApache 2.01.6K 16Jun 2026- Same category
- Commercial use OK
MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradigm Zhang Li, Yuliang Liu, Qiang Liu, Zhiyin Ma, Ziyang Zhang, Shuo Zhang, Zidun Guo, Jiarui Zhang, Xinyu Wang, Xiang Bai
1.2BUnknown licence94 33Jul 2025- Same category
- Smaller (1.2B vs 3.3B)
PP-OCRv6 is a lightweight OCR system that combines architectural innovation with data-centric optimization. It redesigns the backbone, detection neck, and recognition neck around a unified MetaFormer-style building block with structural reparameterization.
?BApache 2.01.4K 19Jun 2026- Same category
- Commercial use OK
Nemotron OCR v2 is a state-of-the-art multilingual text recognition model designed for robust end-to-end optical character recognition (OCR) on complex real-world images.
?BNVIDIA Open Model License2K 272Apr 2026- Same category
- Commercial use OK
korean_PP-OCRv5_mobile_rec is one of the PP-OCRv5_rec that are the latest generation text line recognition models developed by PaddleOCR team. It aims to efficiently and accurately support the recognition of Korean. The key accuracy metrics are as follow:
?BApache 2.07.5K 20Jul 2025- Same category
- Commercial use OK
A higher-precision layout area localization model trained on a self-built dataset containing Chinese and English papers, PPT, multi-layout magazines, contracts, books, exams, ancient books and research reports using RT-DETR-L.
?BApache 2.09K 19Jun 2025- Same category
- Commercial use OK
This Hub repository contains a HuggingFace's transformers implementation of Florence-2 model from Microsoft.
770M1K ctx MIT28.2K 400Jun 2024- Same category
- Smaller (770M vs 3.3B)
- Commercial use OK
General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model - HF Transformers 🤗 implementation
561MApache 2.0136.2K 244Nov 2024- Same category
- Smaller (561M vs 3.3B)
- Commercial use OK
This model is the pretrained version of Pix2Struct, use this model for fine-tuning purposes only.
1.3BApache 2.0616 41Mar 2023- Same category
- Smaller (1.3B vs 3.3B)
- Commercial use OK
This Hub repository contains a HuggingFace's transformers implementation of Florence-2 model from Microsoft.
232M1K ctx MIT20.5M 443Jun 2024- Same category
- Smaller (232M vs 3.3B)
- Commercial use OK
- Widely used
This is the model card of a 🤗 transformers model that has been pushed on the Hub. It has been automatically generated.
823M1K ctx Unknown licence3.8K 65Jun 2024- Same category
- Smaller (823M vs 3.3B)
PP-OCRv6 is a lightweight OCR system that combines architectural innovation with data-centric optimization. It redesigns the backbone, detection neck, and recognition neck around a unified MetaFormer-style building block with structural reparameterization.
19MApache 2.051.4K 35Jun 2026- Same category
- Smaller (19M vs 3.3B)
- Commercial use OK
The Nemotron OCR v1 model is a state-of-the-art text recognition model designed for robust end-to-end optical character recognition (OCR) on complex real-world images.
?BNVIDIA Open Model License233 120Oct 2025- Same category
- Commercial use OK
Granite Docling is a multimodal Image-Text-to-Text model engineered for efficient document conversion. It preserves the core features of Docling while maintaining seamless integration with DoclingDocuments to ensure full compatibility.
258M8K ctx Apache 2.0421.3K 1.3KMay 2025- Same category
- Smaller (258M vs 3.3B)
- Commercial use OK
Indra OCR is a 1M-parameter OCR model that reads up to 49 pages per second on CPU. It turns PDFs and images into markdown, plain text, or JSON with boxes, and it has these features:
1MOpenRAIL6.2K 539Oct 2026- Same category
- Smaller (1M vs 3.3B)
PP-OCRv6 is a lightweight OCR system that combines architectural innovation with data-centric optimization. It redesigns the backbone, detection neck, and recognition neck around a unified MetaFormer-style building block with structural reparameterization.
5MApache 2.012.4K 17Jun 2026- Same category
- Smaller (5M vs 3.3B)
- Commercial use OK