OvisOCR3 is a unified 0.8B multimodal model for OCR and document understanding. It integrates document parsing, general OCR, key information extraction, document figure captioning, and cross-page table reconstruction into a single compact model.
Alternatives to OvisOCR3
120 open-weight models that do the same job as OvisOCR3 (853M 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.
Compare OvisOCR3 with the top 3It works on a range of documents (see usage and benchmarks).
686M256K ctx OpenRAIL668.5K 147May 2026- Same category
- Similar size (686M vs 853M)
Best OCR model . LightOnOCR-2-1B is LightOn's flagship OCR model, refined with RLVR training for maximum accuracy. We recommend this variant for most OCR tasks.
1B16K ctx Apache 2.0726.3K 852Jan 2026- Same category
- Similar size (1B vs 853M)
- Commercial use OK
We also ship smaller models for line-level text detection and ocr error detection. It works on a range of documents (see usage and benchmarks).
689M256K ctx OpenRAIL129.8K 54Oct 2026- Same category
- Similar size (689M vs 853M)
- Newer release
Smallest model. LightOnOCR-3-0.8B is the lightest model of LightOn's LightOnOCR-3 family, with the same layout and visual understanding features.
853M256K ctx Apache 2.015.1K 105Oct 2026- Same category
- Similar size (853M vs 853M)
- Commercial use OK
We are pleased to announce the release of OvisOCR2, a compact 0.8B end-to-end model for page-level document parsing. Given a document page image, OvisOCR2 generates a Markdown representation in natural reading order, covering text, formulas, tables, and visual regions.
853M256K ctx Apache 2.025K 504Jul 2026- Same category
- Similar size (853M vs 853M)
- Commercial use OK
PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training
959M128K ctx Apache 2.099.7K 596May 2026- Same category
- Similar size (959M vs 853M)
- Commercial use OK
MiniCPM-OCR 1.0 (1.3B) is the first OCR-specific model in the MiniCPM-V/o family. It focuses on text-centric document understanding — text, tables, formulas, charts, seals, and structured information extraction — and balances quality against speed on real-world captures.
1.3B256K ctx Apache 2.0243.4K 459Oct 2026- Same category
- Similar size (1.3B vs 853M)
- Commercial use OK
GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture. It introduces Multi-Token Prediction (MTP) loss and stable full-task reinforcement learning to improve training efficiency, recognition accuracy, and generalization.
1.3B128K ctx MIT1.9M 2.2KJan 2026- Same category
- Similar size (1.3B vs 853M)
- Commercial use OK
- Widely used
📦 Model layout. This repository hosts HunyuanOCR-1.5 checkpoint at the root (target base weights). The DFlash speculative-decoding draft lives under dflash/, and the previous HunyuanOCR-1.0 is archived under v1.0/ (load it with subfolder="v1.0", or download the v1.0/ directory directly).
1.1B128K ctx Tencent Hunyuan Community License617.7K 848Nov 2025- Same category
- Similar size (1.1B vs 853M)
PaddleOCR-VL-1.5: Towards a Multi-Task 0.9B VLM for Robust In-the-Wild Document Parsing
959M128K ctx Apache 2.021.9K 668Jan 2026- Same category
- Similar size (959M vs 853M)
- Commercial use OK
Same architecture as LightOnOCR-2-1B, new capabilities. LightOnOCR-3-1B keeps the architecture of LightOn's previous LightOnOCR-2 models, so switching requires no change, and adds the new grounding, image description and chart extraction features.
1B16K ctx Apache 2.04.4K 33Oct 2026- Same category
- Similar size (1B vs 853M)
- Commercial use OK
MinerU2.5-Pro is our latest document parsing model (PDF-to-Markdown) that establishes a new industry standard. By focusing entirely on data engineering without altering the original 1.2B-parameter architecture, it delivers exceptional results across the board:
1.2B8K ctx Apache 2.035.2K 117May 2026- Same category
- Similar size (1.2B vs 853M)
- Commercial use OK
MinerU2.5-Pro is our latest document parsing model (PDF-to-Markdown) that establishes a new industry standard. By focusing entirely on data engineering without altering the original 1.2B-parameter architecture, it delivers exceptional results across the board:
1.2B8K ctx Apache 2.027.9K 185Apr 2026- Same category
- Similar size (1.2B vs 853M)
- Commercial use OK
PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model
959M128K ctx Apache 2.017.2K 1.7KOct 2025- Same category
- Similar size (959M vs 853M)
- Commercial use OK
MinerU2.5 is a 1.2B-parameter vision-language model for document parsing that achieves state-of-the-art accuracy with high computational efficiency.
1.2B16K ctx AGPL 3.077.6K 374Sept 2025- Same category
- Similar size (1.2B vs 853M)
- Commercial use OK
This Hub repository contains a HuggingFace's transformers implementation of Florence-2 model from Microsoft.
777M4K ctx MIT368.6K 1.9KJun 2024- Same category
- Similar size (777M vs 853M)
- Commercial use OK
TeleOCR is a lightweight (~1.2B parameters), open-source Vision-Language Model designed specifically for document parsing.
1.4B125K ctx Apache 2.018.7K 1.3KAug 2026- Same category
- Similar size (1.4B vs 853M)
- Commercial use OK
Base model for fine-tuning. This is the pre-RLVR checkpoint with strong OCR capabilities, ideal as a starting point for domain adaptation and custom fine-tuning.
1B16K ctx Apache 2.03.6K 14Jan 2026- Same category
- Similar size (1B vs 853M)
- Commercial use OK
Best model with image localization. This model predicts both text and precise bounding boxes for embedded images, refined with RLVR using IoU-based rewards for maximum localization accuracy.
1B16K ctx Apache 2.02.9K 30Jan 2026- Same category
- Similar size (1B vs 853M)
- Commercial use OK
General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
716M32K ctx Apache 2.0125.1K 1.6KSept 2024- Same category
- Similar size (716M vs 853M)
- Commercial use OK
Full BF16 version of the model. We recommend this variant for inference and further fine-tuning.
1.2B8K ctx Apache 2.014K 256Oct 2025- Same category
- Similar size (1.2B vs 853M)
- Commercial use OK
Merged bbox variant (advanced). This model combines OCR-improving RLVR signals with bounding-box-focused RLVR updates via joint merging, preserving OCR quality while providing image localization.
1B16K ctx Apache 2.01.5K 18Jan 2026- Same category
- Similar size (1B vs 853M)
- Commercial use OK
This Hub repository contains a HuggingFace's transformers implementation of Florence-2 model from Microsoft.
770M1K ctx MIT28.5K 399Jun 2024- Same category
- Similar size (770M vs 853M)
- Commercial use OK
General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model - HF Transformers 🤗 implementation
561MApache 2.0129.8K 244Nov 2024- Same category
- Similar size (561M vs 853M)
- Commercial use OK