Learn About OpenGradient

OpenGradient Architecture

OpenGradient is an EVM-compatible blockchain network that extends standard blockchain capabilities with native AI inference. The network's novel Hybrid AI Compute Architecture (HACA) leverages specialized node types to provide scalable, verifiable, and decentralized AI inference:

• Full Nodes: Blockchain validator nodes that execute EVM transactions, validate AI inference precompiles, and maintain blockchain state
• Inference Nodes: Specialized validator nodes with optimized hardware (CPUs, GPUs, TPUs, TEEs) that execute AI model inference and generate verification proofs
• Storage Nodes: Decentralized storage infrastructure powered by Walrus that maintains the Model Hub's filestore, storing model files referenced by Blob ID
• Data Nodes: Secure enclave nodes that provide trusted data access services for smart contracts, enabling encrypted connections to third-party data sources

Learn more about OpenGradient's unique architecture and how these components work together in OpenGradient Architecture.

Verifiable AI Execution

OpenGradient supports two execution modes for AI inference:

• ML Execution (PIPE): Native smart contract execution with pre-execution inference, enabling ML models to be called directly from EVM transactions with ZKML, TEE, or Vanilla verification
• LLM Execution (x402): Payment-gated HTTP API execution using the HTTP 402 Payment Required standard with TEE verification, ideal for web applications and LLM inference