OpenDataLab provides access to numerous significant open-source datasets.
67
Public repositories
90,614
Total stars
2,785
Followers
OpenDataLab maintains a significant public presence on GitHub, showcasing a variety of repositories primarily in Python and TypeScript. Notable projects include MinerU, a tool for transforming complex documents, and PDF-Extract-Kit, which offers high-quality PDF content extraction solutions. The organization's focus on open-source datasets enhances accessibility and collaboration in data science.
Top languages
Public repositories
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
PDF-Extract-Kit
A Comprehensive Toolkit for High-Quality PDF Content Extraction
DocLayout-YOLO
DocLayout-YOLO: Enhancing Document Layout Analysis through Diverse Synthetic Data and Global-to-Local Adaptive Perception
OmniDocBench
[CVPR 2025] A Comprehensive Benchmark for Document Parsing and Evaluation
labelU
Open-source multimodal data annotation platform with AI auto-annotation support.
LabelLLM
The Open-Source Data Annotation Platform
MinerU-Diffusion
A diffusion-based framework for document OCR that replaces autoregressive decoding with block-level parallel diffusion decoding.
MinerU-Document-Explorer
Agent-native knowledge engine with MCP tools for document indexing, wiki organization, fast retrieval and deep reading across PDF/DOCX/PPTX/Markdown
WanJuan1.0
万卷1.0多模态语料
magic-doc
No description provided for this repository.
magic-html
No description provided for this repository.
UniMERNet
UniMERNet: A Universal Network for Real-World Mathematical Expression Recognition
MinerU-HTML
MinerU-HTML: An SLM-powered HTML main content extractor that outputs clean HTML bodies. Perfect for Deep Research Agents, RAG applications, and training data generation.
Meta-rater
[ACL 2025 Best Theme Paper] This is the official implementation for the paper: "Meta-rater: A Multi-dimensional Data Selection Method for Pre-training Language Models"
LOKI
[ICLR 2025 Spotlight] The official implementation of the paper “LOKI：A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Models”
MinerU-Popo
No description provided for this repository.
Earth-Agent
[ICLR 2026] The official implementation of the paper “Earth-Agent: Unlocking the Full Landscape of Earth Observation with Agents”
labelU-Kit
Data annotation component library --provided as NPM packages
FakeVLM
[NeurIPS 2025 🔥] FakeVLM: Advancing Synthetic Image Detection through Explainable Multimodal Models and Fine-Grained Artifact Analysis
opendatalab-datasets
datasets resource
mineru-vl-utils
A Python package for interacting with the MinerU Vision-Language Model.
laion5b-downloader
No description provided for this repository.
MinerU-Ecosystem
No description provided for this repository.
VHM
VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis
HA-DPO
Beyond Hallucinations: Enhancing LVLMs through Hallucination-Aware Direct Preference Optimization
OHR-Bench
(ICCV 2025) OCR Hinders RAG: Evaluating the Cascading Impact of OCR on Retrieval-Augmented Generation
VIGC
AAAI 2024: Visual Instruction Generation and Correction
MLS-BRN
[CVPR 2024] 3D Building Reconstruction from Monocular Remote Sensing Images with Multi-level Supervisions
Vis3
Data browser based on s3. 一个基于 S3 的数据（json / jsonl / parquet / html / md等）可视化工具。👇 Try online.
skydiffusion
[ICCV 2025] The official implementation of the paper “Street-to-Satellite Image Synthesis with Diffusion Models and BEV Paradigm”
LEGION
[ICCV25 Highlight] The official implementation of the paper "LEGION: Learning to Ground and Explain for Synthetic Image Detection"
CiteVQA
No description provided for this repository.
CLIP-Parrot-Bias
ECCV2024_Parrot Captions Teach CLIP to Spot Text
opendatalab-python-sdk
SDK of OpenDataLab - https://opendatalab.org.cn
MLLM-DataEngine
MLLM-DataEngine: An Iterative Refinement Approach for MLLM
WanJuan3.0
WanJuan3.0（“万卷·丝路”）一个作为综合性的纯文本语料库，采集了多个国家地区的网络公开信息、文献、专利等资料，数据总规模超1.2TB，Token总数超过300B，处于国际领先水平，首期开源的语料库主要由泰语、俄语、阿拉伯语、韩语和越南语5个子集构成，每个子集的数据规模均超过150GB
dsdl-docs
Data Set Description Language Specification （新一代人工智能数据集描述语言DSDL）
ProverGen
[ICLR 2025] This is the official implementation for the paper: "Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluation"
CHARM
[ACL 2024 Main Conference] Chinese commonsense benchmark for LLMs
UrBench
[AAAI 2025]This repo contains evaluation code for the paper “UrBench: A Comprehensive Benchmark for Evaluating Large Multimodal Models in Multi-View Urban Scenarios”
TRivia
(CVPR 2026) TRivia: Self-supervised Fine-tuning of Vision-Language Models for Table Recognition
REST
No description provided for this repository.
image-downloader
No description provided for this repository.
Sciverse-Agent-Tools
Standardized tool schemas and SDKs that expose Sciverse Open Platform retrieval capabilities to LLM agents.
RxnCaption
[CVPR 2026] SOTA Chemical Reaction Diagram Parsing Framework
labelbee
No description provided for this repository.
Miner-PDF-Benchmark
MPB (Miner-PDF-Benchmark) is an end-to-end PDF document comprehension evaluation suite designed for large-scale model data scenarios.
mineru-tutorials
MinerU Training Camp course materials and tutorials
awesome-markdown-ebooks
No description provided for this repository.
WebMainBench
WebMainBench is a high-precision benchmark for evaluating web main content extraction.
CrossViewDiff
The official implementation of the paper "CrossViewDiff: A Cross-View Diffusion Model for Satellite-to-Street View Synthesis"
WanJuan2.0-WanJuan-CC
WanJuan-CC是以CommonCrawl为基础，经过数据抽取，规则清洗，去重，安全过滤，质量清洗等步骤得到的高质量数据。
PM4Bench
No description provided for this repository.
dsdl-sdk
No description provided for this repository.
OpenHuEval
No description provided for this repository.
MinerU-Webkit
No description provided for this repository.
labelU-frontend
LabelU front-end library
awesome-mineru
🕶️ A curated list of awesome things related to MinerU
allz
A universal command line tool for compression and decompression
MolRecBench-Wild
No description provided for this repository.
CRaFT
[AAAI25] Certainty Represented Knowledge Flow for Refusal-Aware Instruction Tuning
GRAIT
[NAACL25 findings] Gradient-Driven Refusal-Aware Instruction Tuning for Effective Hallucination Mitigation
OmniDocLayout
[CVPR26 Highlight] The official implementation of the paper "OmniDocLayout: Towards Diverse Document Layout Generation via Coarse-to-Fine LLM Learning"
.github
No description provided for this repository.
opendatalab.github.io
No description provided for this repository.
labelU-ML
No description provided for this repository.
rdkit
A forked repo of the official RDKit library
Frequently asked questions
What does opendatalab build on GitHub?
OpenDataLab develops a range of projects on GitHub, focusing on open-source datasets and tools for document processing. Their notable repositories include MinerU and PDF-Extract-Kit, which cater to various data extraction and analysis needs.
Which programming languages does opendatalab use?
OpenDataLab primarily utilizes Python and TypeScript for their projects, along with HTML, JavaScript, and Jupyter Notebook. This diverse set of languages supports their various initiatives in document processing and data annotation.
Are opendatalab's repositories public?
Yes, all of OpenDataLab's repositories are public on GitHub. This transparency allows users to access their projects, collaborate, and contribute to the development of tools focused on open-source datasets and document processing.
Is this exposure intended?
Monitor OpenDataLab with RepoGuard and get alerted the moment a new public repository appears.
Monitor this account