Monogmos AI

Monogmos AI – The Decentralized AI Oracle for Smart Contracts. Bringing verifiable AI-powered decisions to the NEAR Protocol with trustless execution. 🚀

Comment

Monogmos AI

Introduction

Monogmos AI is a decentralized AI oracle designed to seamlessly integrate AI functionalities into smart contracts on the NEAR Protocol. By leveraging a network of permissionless nodes, Monogmos enables smart contracts to make AI-powered decisions in a trustless and verifiable manner.

With Monogmos AI, developers can incorporate OpenAI LLM requests into their smart contracts without relying on centralized infrastructure, paving the way for decentralized AI applications on the blockchain.

Inspiration

The growing demand for AI-powered blockchain applications has highlighted the need for decentralized, verifiable AI oracles. Traditional AI integrations in Web3 rely on centralized services, limiting transparency and trust. Monogmos AI was created to solve this problem—bringing AI-driven decisions to smart contracts in a fully decentralized manner.

What It Does

Monogmos AI is a decentralized AI oracle designed to seamlessly integrate AI capabilities into NEAR smart contracts. It enables developers to:

  • Run AI-powered smart contracts on NEAR, aligning with the Bitte NEAR/EVM Chain Agent Bounty.
  • Deploy AI-driven blockchain agents that can evolve and make decisions, fitting the Aurora: AI Agentic Chains Bounty.
  • Automate cross-chain AI interactions, contributing to the Proximity Labs Cross-Chain Trading Agents Bounty.

How We Built It

  • Smart Contracts – Developed on the NEAR Protocol for AI-powered interactions, making it a contender for the Frax Finance Integration Bounty.
  • Oracle Network – Utilized ZeroMQ and Socket.IO to distribute AI tasks, supporting Nuffle Labs’ best integration challenge.
  • AI Integration – Used OpenAI’s LLMs for function calling and automation, aligning with the VeaxFlow AI Agent Bounty for liquidity management AI.
  • Frontend & Backend – Built with Next.js and Express for seamless developer access, enhancing usability for projects like Questflow’s AI-powered developer tools.

Challenges We Ran Into

  • Decentralizing AI computation – Ensuring AI tasks are processed transparently and verifiably.
  • Latency vs. Security – Balancing fast AI responses with trustless verification.
  • Seamless Blockchain Integration – Ensuring smooth cross-chain AI operations, key for Proximity Labs’ Cross-Chain Trading Agents Bounty.

Accomplishments That We’re Proud Of

  • Developed a working decentralized AI oracle that enables verifiable AI-powered decisions.
  • Implemented multi-agent AI coordination for trading and automation, advancing Swanchain’s Marketing Agents challenge.
  • Built an AI-powered contract compatible with HOT Wallet SDK and Omni bridge, making it eligible for HOT Wallet Bounty credits.

What We Learned

  • AI verifiability is crucial – Zero-Knowledge Proofs can enhance blockchain AI trust.
  • Optimizing AI for real-time trading – Key for Proximity Labs Bitcoin & Trading Agents bounty.
  • Marketing AI agents can bridge Web2 and Web3 – Valuable for Swanchain’s Marketing Agents Bounty.

What's Next for Monogmos AI

🚀 Zero-Knowledge AI Proofs – Bringing cryptographic verification to AI outputs.
📡 AI-Powered Trading & Data Agents – Strengthening Questflow’s Airdrop & Strategic Trading Agents.
📈 Expanded AI Model Support – Including multimodal AI (text, vision, voice) for broader applications.
🔗 Cross-Chain AI Oracles – Enhancing Aurora-based agentic AI chains for multi-chain AI deployment.

Features

  • Decentralized AI Computation – A distributed network of nodes processes AI tasks, ensuring trustless execution.
  • Transparent & Verifiable Results – AI-generated outputs are recorded on-chain for full auditability.
  • Scalable & Efficient – Multiple compute requests are handled in parallel to optimize performance.

Getting Started

Integration with NEAR Smart Contracts

To integrate Monogmos AI into your NEAR smart contract, follow our detailed guide:
📖 Documentation

Demo & Code Examples

Simple GPT Integration

Function Calling with OpenAI

Agent Contract

Deployment Addresses

Contract Type Account Address
Oracle Contract oracletest2.testnet
Function Calling Contract functioncalltest1.testnet
Agent Contract agentstaging.testnet
SimpleGPT Contract simplegpttest.testnet

Technology Stack

Monogmos AI is built using:

  • NEAR Protocol’s Events Indexer – For tracking and indexing contract interactions.
  • Socket.IO – For real-time communication between components.
  • ZeroMQ – For message queuing and task distribution.
  • NEAR CLI – For smart contract deployment and management.
  • Next.js & Express – Powering the frontend and backend services.

Inspiration

The intersection of AI and blockchain unlocks new possibilities for verifiable and decentralized AI applications. Monogmos AI aims to bridge this gap by providing an oracle that enables AI-powered smart contracts, enhancing trust, transparency, and scalability in the Web3 space.

Introduction

Monogmos AI serves as a decentralized AI oracle designed to empower developers by seamlessly integrating powerful AI functionalities into any smart contract on the NEAR Protocol. This innovative platform allows any NEAR smart contract to call the Monogmos oracle contract, enabling it to leverage advanced AI capabilities without the constraints of traditional centralized solutions. By utilizing a network of permissionless nodes, Monogmos AI processes complex OpenAI LLM requests, returning transparent and verifiable results. This integration not only enhances the functionality of smart contracts but also paves the way for a new era of decentralized AI applications on the blockchain, where trust, scalability, and efficiency are paramount.

mindmap

Documentation

Here is a detailed documentation to integrate Monogmos AI Oracle into your Near SmartContract. Documentation

Demo

Simple GPT

FunctionCall in OpenAi

Agent

Deployment addresses

  • Oracle contract account: oracletest2.testnet
  • Function calling contract account: functioncalltest1.testnet
  • Agent contract account: agentstaging.testnet
  • SimpleGPT contract account: simplegpttest.testnet

Inspiration

At the intersection of AI and crypto lies untapped potential. Many web3 products envision on-chain AI as the future, bringing transparency, provability, and decentralization to AI interactions. Monogmos AI addresses the limitations of existing oracle solutions by enabling verifiable and decentralized AI computation on-chain.

Features

  • Decentralized AI Computation: Utilize a network of permissionless nodes to run AI models directly on-chain.
  • Transparent Results: Receive detailed outputs from AI computations that can be audited for integrity.
  • Scalability: Handle multiple compute requests in parallel, ensuring efficient processing.

How We Built It

We utilized a range of cutting-edge tools to develop Monogmos AI, including:

  • NEAR Protocol’s Events Indexer: For indexing events and interactions.
  • Socket.IO: For real-time communication between components.
  • ZeroMQ: To facilitate message queuing and processing.
  • NEAR CLI: For deploying and managing our smart contracts.
  • Next.js & Express: For building the frontend and backend of the application.

Key Accomplishments

  • Implemented a distribution layer for our oracle service, enabling multiple Monogmos nodes to handle compute requests in parallel, which maximizes efficiency in a decentralized environment.

Roadmap

What's next for Monogmos AI:

  • Zero-Knowledge Verifiability: Implement cryptographic proofs for AI computations to enhance trust and reliability.
  • Enhanced Data Primitives: Develop richer data structures to better support our AI initiatives.
  • Incentivize the Nodes: Create a rewards system for nodes that contribute computational power by connecting to Monogmos.

Built With

Share this project:

Updates