> For the complete documentation index, see [llms.txt](https://cluster-protocol.gitbook.io/whitepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cluster-protocol.gitbook.io/whitepaper/abstract.md).

# Abstract

**Cluster Protocol** is a unified AI infrastructure layer for the decentralized web and an orchestration layer for autonomous workflows. It consolidates three foundational primitives - **AI** **model inference, tokenized data, and compute,** into a single orchestration platform with one API, one settlement layer, and **singular AI economy on Base (Ethereum L2)**.

The platform provides an AI inference gateway serving 500+ AI models across multiple modalities (text, image, audio, embeddings, reranking), a tokenized data marketplace where datasets are stored on IPFS and ownership is represented as on-chain NFTs with automatic revenue distribution, and a compute layer that provisions GPU resources for custom model hosting and fine-tuning workloads.

Payments across the stack are settled natively on Base via smart contracts, with support for x402, an HTTP-native micropayment protocol that enables AI agents and developers to pay per request without accounts, API keys, or pre-deposited balances.

Coupled with a browser based IDE, an prompt-to-dApp application layer built on top of Cluster's infrastructure, enables anyone to create, deploy, and distribute fully on-chain applications using natural language, consuming Cluster inference, data, and compute natively in the process.

This paper describes the architecture, components, and mechanisms of Cluster Protocol.