OpenGraph Explorer

Bringing Transparency to AI through On-Chain Machine Learning

SUI Overflow !!

Our Overflow Deck [Overflow, Decentralized Data Annotation Platform]: https://www.figma.com/deck/jz6NCLwVX3RazYHqBvoOnE/OpenGraph--Overflow2025--AI-?node-id=6-1055&t=ZetcuAb0izQp9lVz-0&scaling=min-zoom&content-scaling=fixed&page-id=0%3A1

Vote! [AI Track - OpenGraph] https://vote.sui.io/

Twitter : https://x.com/opengraph_labs

🌟 Vision

OpenGraph Explorer is pioneering a new era of transparent and decentralized AI by bringing machine learning models fully on-chain. We're transforming the traditional "black box" nature of AI into a transparent, auditable, and community-owned ecosystem powered by the Sui blockchain.

Our platform allows developers to deploy ML models on-chain and execute inference operations with complete visibility - tracking every layer, every calculation, and every transformation directly on the blockchain. This creates an unprecedented level of transparency and trust in AI systems.

✨ Key Features

• On-Chain Model Repository: Deploy your ML models directly to the Sui blockchain
• Transparent Layer-by-Layer Inference: Witness and verify every step of the inference process
• Real-Time Execution Tracking: Monitor each layer's processing status as inference occurs
• Blockchain Verification: Every operation is recorded on-chain with transaction receipts
• Visual Model Exploration: Interactive UI for exploring model architecture and inference results
• Multi-Format Support: Compatible with various model formats with automatic conversion for on-chain deployment

🔍 Why On-Chain Machine Learning?

Traditional ML/AI systems suffer from opacity - users must trust black-box models without visibility into their operations. OpenGraph Explorer solves this by:

• Complete Transparency: Every model parameter and inference calculation is visible and verifiable on-chain
• Auditability: The entire model execution path can be audited by following on-chain transactions
• Immutability: Model architectures and parameters are immutably recorded on the blockchain
• Decentralization: No central authority controls model access or execution
• Community Ownership: Models become public goods that anyone can access, verify, and build upon

🖥️ Technology Stack

• Frontend: React, TypeScript, Vite, Radix UI
• Blockchain: Sui Network
• Wallet Connection: @mysten/dapp-kit
• Visualization: Custom layer-by-layer inference visualization
• Styling: CSS Modules, Framer Motion
• Package Management: Yarn/npm

🏗️ Architecture

OpenGraph Explorer breaks down complex ML models into layers and executes them sequentially on the Sui blockchain:

1. Model Upload & Decomposition: Models are decomposed into their constituent layers
2. On-Chain Deployment: Each layer is stored as an immutable object on the Sui blockchain
3. Layer-by-Layer Execution: Input vectors propagate through each layer with full transparency
4. Transaction Verification: Each layer execution generates a verifiable transaction record
5. Real-Time Visualization: The UI provides a visual representation of the entire process

🚀 Getting Started

Prerequisites

• Node.js (v16 or later)
• Yarn/npm package manager
• A Sui wallet (like Sui Wallet browser extension)

Installation

1. Clone the repository:

   git clone https://github.com/OpenGraphLabs/opengraph-explorer.git
   cd opengraph-explorer

2. Install dependencies:

   yarn install
   # or
   npm install

3. Start the development server:

   yarn dev
   # or
   npm run dev

4. Open your browser and navigate to http://localhost:5173

🔄 Workflow

1. Connect Wallet: Connect your Sui wallet to access the platform
2. Explore Models: Browse through available on-chain models
3. Upload Models: Deploy your ML models to the Sui blockchain
4. Execute Inference: Run transparent, layer-by-layer inference with any input
5. Verify Results: Follow each step of the execution with on-chain verification
6. Analyze Outputs: Examine final outputs and the entire execution path

🌐 Global Impact

OpenGraph Explorer is designed for the global AI community, enabling:

• Researchers: Verify model behaviors and compare execution patterns across models
• Developers: Build applications on top of trusted, transparent AI models
• Auditors: Analyze models for bias, security vulnerabilities, or unexpected behaviors
• End Users: Gain confidence in AI systems through unprecedented transparency

🔮 Future Roadmap

• Model Composition: Combine multiple on-chain models to create new architectures
• Federated Training: Distribute model training across the network
• Governance Mechanisms: Community-driven decision making for platform development
• Performance Optimization: Enhance on-chain execution efficiency
• Expanded Model Support: Additional model architectures and layer types
• Cross-Chain Integration: Extend to other blockchain networks

🛠️ Project Structure

opengraph-explorer/
├── client/
│   ├── src/
│   │   ├── components/
│   │   │   ├── model/
│   │   │   │   ├── ModelInferenceTab.tsx
│   │   │   │   ├── ModelOverviewTab.tsx
│   │   │   │   └── ...
│   │   ├── pages/
│   │   │   ├── Home.tsx
│   │   │   ├── Models.tsx
│   │   │   ├── ModelDetail.tsx
│   │   │   └── ...
│   │   ├── utils/
│   │   │   ├── modelUtils.ts
│   │   │   ├── sui.ts
│   │   │   └── ...
│   │   ├── App.tsx
│   │   └── main.tsx
├── server/ (Optional backend services)
└── README.md

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request or open an Issue for discussion.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgements

• Sui Network for blockchain infrastructure
• Mysten Labs for Sui ecosystem tools
• All contributors and community members

---

Built with ❤️ by OpenGraph Labs

Twitter • Discord • GitHub