OpenGradient Docs

Appearance

Developers Overview ​

OpenGradient provides three main developer tools for building AI-enabled applications on top of OpenGradient infrastructure, each designed for specific distinct use cases: SolidML, OpenGradient SDK, and MemSync.

SolidML ​

SolidML is a Solidity framework for building AI-enabled smart contracts and on-chain applications on the OpenGradient network. It enables you to run ML and LLM inference directly from your smart contracts as part of atomic transactions. Key features include:

TIP

Use SolidML if you're building dApps, protocols, or any on-chain applications that need AI capabilities.

Python SDK ​

Python SDK enables you to integrate OpenGradient's decentralized, verified AI infrastructure into traditional off-chain applications. The SDK abstracts away blockchain complexity while maintaining full security and decentralization. Key features include:

  • ML Inference & Workflows: Run ML model inference and deploy automated workflows with scheduled execution entirely onchain
  • LLM Inference: Use secure and verifiable large language models for completions and chat with support for tools and TEE execution
  • Model Management: Upload, manage, and organize models on the Model Hub

TIP

Use the Python SDK if you're building traditional applications (web apps, APIs, scripts) that need decentralized AI capabilities.

MemSync ​

MemSync is a long-term memory layer for AI built on top of OpenGradient's verifiable inference and embeddings infrastructure. It provides a REST API for long-term context management and AI personalization. Key features include:

  • Fact Extraction: Extract and store semantic (long-term) and episodic (temporary) facts from conversations
  • Semantic Search: Search across personal memories and context using semantic similarity
  • User Profiles: Manage user profiles and preferences for personalized AI experiences
  • Context Enrichment: Integrate with external services for enhanced context

TIP

Use MemSync if you're building AI applications that need persistent memory, user personalization, or long-term context management.

Security & Decentralization ​

SolidML, the Python SDK, and MemSync are powered by the OpenGradient network and provide the same security guarantees, including:

  • Decentralization: No single point of failure
  • Censorship resistance: Open access to AI models
  • Verifiability: Inferences are cryptographically verified
  • Transparency: Full auditability of model execution