# Overview

OpenShard is a private, decentralized AI inference network built on Base. It enables compute providers — referred to as sellers — to expose model-serving endpoints, while client-side gateways — referred to as buyers — discover those peers, route private inference requests directly to them peer-to-peer, and attach per-request micropayment authorization data through the x402 protocol. No centralized intermediary or single trusted party is required to broker any inference transaction at any point in the network.

The project is built around three primary engineering concerns: decentralized service discovery, privacy-preserving request transport, and on-chain micropayment economics native to Base. Privacy-preserving transport is enforced through a layered stack of cryptographic mechanisms — Zero-Knowledge proofs allow buyers to prove credential validity without disclosing account identity, End-to-End Encryption keeps inference content sealed in transit from buyer to seller, and Trusted Execution principles ensure private keys never leave the process boundary they are assigned to. These are implemented as a Node.js monorepo with independent packages for each protocol role, allowing teams to run individual components in isolation or as a coordinated network.

Three design principles shape every layer of OpenShard. Compatibility-first integration means buyers interact with standard OpenAI and Anthropic-style request shapes, so existing applications require minimal changes to route inference through the network. Trust minimization means no component in the data path needs to be trusted unconditionally — cryptographic verification replaces institutional trust at each boundary between buyer, seller, and registry. Composable deployment means services can run standalone, as part of a multi-node simulation, or embedded inside integration test harnesses, making the architecture suitable for individual operators, researchers, and production-scale network deployments alike.

<br>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.openshard.ai/getting-started/overview.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.