OpenCodePapers

optical-character-recognition-ocr-on-ocrbench

Optical Character Recognition (OCR)
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeScoreModelNameReleaseDate
Qwen3.5: Towards Native Multimodal Agents✓ Link931Qwen3.5-397B-A17B2026-02-16
Kimi K2.5: Visual Agentic Intelligence✓ Link923Kimi K2.52026-02-02
Qwen3-VL Technical Report✓ Link920Qwen3-VL-235B-A22B-Instruct2025-11-26
SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture✓ Link919SenseNova-U1-A3B-MoT2026-05-12
Qwen3.5-Omni Technical Report913Qwen3.5-Omni-Plus2026-04-21
OpenVLThinkerV2: A Generalist Multimodal Reasoning Model for Multi-domain Visual Tasks✓ Link911.0OpenVLThinkerV2 (Ours)2026-04-09
InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models✓ Link906InternVL3-78B2025-04-14
Qwen3.6✓ Link900Qwen3.6-35B-A3B2026-04-21
Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders✓ Link896Penguin-VL 8B2026-03-06
Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA✓ Link896Macaron-V1-Tall2026-08-10
MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe✓ Link890MiniCPM-V 4.52025-09-16
DeepAlign: Mitigating Modality Conflict through Modality-Specific Alignment890DeepAlign (InternVL3-8B)2026-06-03
Qwen2.5-VL Technical Report✓ Link885Qwen2.5-VL-72B2025-02-19
Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence✓ Link883Nemotron 3 Nano Omni2026-04-27
Step-GUI Technical Report✓ Link880Step-GUI-8B2025-12-17
Qianfan-OCR: A Unified End-to-End Model for Document Intelligence✓ Link880Qianfan-OCR2026-03-11
Ovis2.5 Technical Report✓ Link879Ovis2.5-9B2025-08-15
ERNIE 5.0 Technical Report878.0ERNIE 5.02026-02-04
Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution✓ Link877Qwen2-VL-72B2024-09-18
Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning✓ Link877FactoSR-8B-SFT2026-09-03
MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction✓ Link876MiniCPM-o 4.5-Instruct2026-04-30
ERNIE 4.5 Technical Report✓ Link872ERNIE-4.5-VL-424B-A47B2025-06-30
PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding✓ Link870PLM-8B2025-04-17
Kimi-VL Technical Report✓ Link869Kimi-VL-A3B-Thinking-25062025-04-10
Introducing Command A Vision: Multimodal AI built for business869Command A Vision2025-07-31
Kwai Keye-VL 1.5 Technical Report✓ Link866Keye-VL-1.5-8B2025-09-01
MiniMax-01: Scaling Foundation Models with Lightning Attention✓ Link865MiniMax-VL-012025-01-14
GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning✓ Link865GLM-4.6V2025-07-01
LongCat-Next: Lexicalizing Modalities as Discrete Tokens✓ Link865.0LongCat-Next (A3B, 68.5B)2026-03-29
Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Driving✓ Link864Qwen-Drive-1.0-SFT2026-08-31
Infinity-Parser2-Pro862.0Infinity-Parser2-Pro2026-05-11
Unlocking Dense Metric Depth Estimation in VLMs✓ Link862DepthVLM-8B2026-05-15
HunyuanOCR-1.5: Making Lightweight OCR VLMs Faster and Better✓ Link861HunyuanOCR-1.5 (1B)2026-07-06
HunyuanOCR Technical Report✓ Link860HunyuanOCR2025-11-24
Multimodal OCR: Parse Anything from Documents✓ Link860dots.mocr2026-03-13
FocusVTC: Efficient and High-Performance Visual Text Compression with Adaptive Resolution✓ Link860FocusVTC2026-09-29
HyperVL: An Efficient and Dynamic Multimodal Large Language Model for Edge Devices859.0HyperVL ViTL2025-12-16
LLaVA-UHD v4: What Makes Efficient Visual Encoding in MLLMs?✓ Link859LLaVA-UHD v42026-05-09
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone857MiniCPM-V 4.6-Thinking (4x)2026-05-13
Kwai Keye-VL-2.0 Technical Report✓ Link857Keye-VL-2.0-30B-A3B2026-06-09
NVIDIA Nemotron Nano V2 VL✓ Link856Nemotron Nano V2 VL (Reasoning off) (extra data)2025-11-06
LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents✓ Link855LLaDA-UI (pre-training)2026-09-09
EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience✓ Link853.5EvoCUA-32B2026-01-22
Kwai Keye-VL Technical Report✓ Link851Keye-VL-8B2025-07-02
Thinking with Images via Self-Calling Agent✓ Link845.0SubagentVL2025-12-09
Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs✓ Link844Phi-4-Multimodal2025-02-24
DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding✓ Link834DeepSeek-VL2-Small2024-12-13
Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification✓ Link833.0UniAR2026-06-16
POINTS1.5: Building a Vision-Language Model towards Real World Applications✓ Link832POINTS1.5-7B2024-12-11
NGM: A Plug-and-Play Training-Free Memory Module for LLMs✓ Link829Qwen3-VL-2B + NGM2026-05-16
VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding✓ Link828VideoLLaMA3 7B2025-01-22
PP-DocBee: Improving Multimodal Document Understanding Through a Bag of Tricks✓ Link828PP-DocBee-2B2025-03-06
SenseNova-U1.5: Towards Native Unified Visual Intelligence✓ Link824.6SenseNova-U1.5 8B-Think2026-09-10
LFM2 Technical Report✓ Link822LFM2-VL-3B2025-11-28
Mage-VL: An Efficient Codec-Native Streaming Multimodal Foundation Model✓ Link818Mage-VL-4B2026-07-27
From Pixels to Words -- Towards Native One-Vision Models at Scale✓ Link816NEO-ov 9B2026-05-27
VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning✓ Link808VisionThink Efficient2025-07-17
Mini-Monkey: Alleviating the Semantic Sawtooth Effect for Lightweight MLLMs via Complementary Image Pyramid✓ Link806Mini-Monkey (extra data)2024-08-04
InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models✓ Link798InfiniteVL-Base (extra data)2025-12-09
ZAYA1-VL-8B Technical Report✓ Link798ZAYA1-VL-8B2026-05-08
Eagle: Exploring The Design Space for Multimodal LLMs with Mixture of Encoders✓ Link788Eagle-X5-34B-Plus2024-08-28
CASA: Cross-Attention via Self-Attention for Efficient Vision-Language Fusion✓ Link783.0CAQ2.5-VL2025-12-22
LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence✓ Link782LLaVA-OneVision-2-8B2026-05-25
From Pixels to Words -- Towards Native Vision-Language Primitives at Scale✓ Link777NEO-9B2025-10-16
Oryx MLLM: On-Demand Spatial-Temporal Understanding at Arbitrary Resolution✓ Link746Oryx-1.5 32B2024-09-19
FastVLM: Efficient Vision Encoding for Vision Language Models✓ Link731FastVLM-7B2024-12-17
SmolVLM: Redefining small and efficient multimodal models✓ Link729SmolVLM 2.2B2025-04-07
MiniCPM-V: A GPT-4V Level MLLM on Your Phone✓ Link725MiniCPM-Llama3-V 2.52024-08-03
How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites✓ Link724InternVL 1.52024-04-25
Uncovering Understanding-Generation Synergy in Native Unified Multimodal Models: From Representation, Task to System689Native UMM Synergy: Dense U+G2026-09-01
Emu3: Next-Token Prediction is All You Need✓ Link687Emu32024-09-27
LFM2.5-VL-450M: Structured Visual Intelligence, Edge to Cloud684LFM2.5-VL-450M2026-04-08
Let ViT Speak: Generative Language-Image Pre-training✓ Link659GenLIP-g/16 + LLaVA-NeXT-Qwen2.5-7B2026-05-01
OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence✓ Link657OV-Encoder-Lang + Qwen3-4B-Instruct25072026-02-09
Florence-VL: Enhancing Vision-Language Models with Generative Vision Encoder and Depth-Breadth Fusion✓ Link634Florence-VL 8B2024-12-05
MonkeyOCRv2: A Visual-Text Foundation Model for Document AI✓ Link581MonkeyOCRv2-B-Und2026-07-13
TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document✓ Link561TextMonkey2024-03-07
DeepSeek-VL: Towards Real-World Vision-Language Understanding✓ Link456DeepSeek-VL 7B Chat2024-03-08
SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion✓ Link338SmolDocling-256M-preview2025-03-14
Check out how to contribute new results. Then edit this file on GitLab or GitHub.