Hero Hero

​
Welcome to Atoma

Atoma is a decentralized private and verifiable AI execution network. It provides the infrastructure for developers and enterprises to use open-source AI models and build AI applications with full privacy. There are several key components to Atoma’s infrastructure:
  • Decentralized AI Compute:
    • Atoma has a global network of GPUs that register on the network to perform inference requests, referred to as execution nodes.
    • Rather than processing requests through a single data centre, Atoma is a fully permissionless network that works with data centres around the world to intelligently distribute AI requests.
  • Dynamic Marketplace:
    • Node operators set rates for their compute resources.
    • Users select nodes based on cost, performance, and privacy preferences.
  • Atoma Proxy:
    • Each request is distributed to the most relevant node based on the hardware spec and whether privacy is requested, for that particular request.
    • This ensures cost efficiency and performance.
  • Privacy and Security:
    • Trusted Execution Environments (TEEs) ensure computations are secure and data remains private.
    • The Sampling Consensus Protocol verifies computation integrity across the network.
    • Data and model weights are inaccessible to unauthorized parties, protecting user and model provider assets.
  • OpenAI API Compatibility:
    • The network supports OpenAI-compatible APIs, allowing developers to integrate decentralized AI capabilities without reworking existing systems.
  • Community Governance:
    • Atoma’s treasury is managed by the community, with transparent reward distribution.

​
Getting Started

Choose your path to get started with Atoma:

Use the API

Access decentralized AI models through our OpenAI-compatible API

Run a Node

Deploy your AI models and earn rewards by contributing compute resources to the network

​
Explore Atoma’s AI Compute Layer

Decentralized Infrastructure

Leverage a distributed network of compute nodes for reliable AI inference

Multiple Model Support

Deploy and access various AI models including LLMs, embeddings, and image generation

Smart Contracts

Secure and transparent handling of compute resources and payments

Developer Tools

Comprehensive SDKs and APIs for seamless integration