> For the complete documentation index, see [llms.txt](https://docs.ama.one/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ama.one/ecosystem/partners/crust-network.md).

# Crust Network

Type: Infrastructure - decentralized storage 

Website: <https://crust.network/>

Crust Network is a decentralized, trustless storage layer built on IPFS and supported by thousands of globally distributed nodes. It provides censorship-resistant, verifiable storage for datasets, model artifacts, logs, and application state — all backed by a market-driven storage economy and TEE-enhanced security guarantees.

Integrated with Amadeus, Crust becomes the off-chain memory layer powering AI agents, oracle streams, model provenance, and long-term dataset storage. This positions Amadeus as a full-stack decentralized AI cloud: compute on Amadeus, storage on Crust.

## **Planned Capabilities for Amadeus**

Once integration is complete, Crust Network will enable Amadeus to:

* **Store AI agent memory, logs, and checkpoints** in decentralized storage, with hashes anchored on-chain for provenance.
* **Extend Oracle Streams** by providing a durable archival layer for historical datasets, enrichments, and inference outputs.
* **Support persistent agent states** across long-running autonomous workflows with tamper-proof storage.
* **Enable decentralized dataset hosting**, allowing developers to provide training data that Amadeus agents can access via IPFS.
* **Provide cross-ecosystem wallet compatibility** (SubWallet, Talisman, Crust Wallet) for seamless user authentication and asset flow.
* **Create a durable data backend for uPoW compute**, keeping training artifacts and model updates accessible to verifiers, enterprises, and permissioned workflows.

## **Example Use Cases**

#### **Decentralized AI Memory**

Amadeus agents persist memory, logs, and model checkpoints on Crust, ensuring tamper-proof lineage and long-term versioning.

#### **Archival Layer for Oracle Streams**

Live data served via Oracle Streams is continuously archived on Crust, enabling agents to learn from historical time-series datasets.

#### **Dataset Hosting for AI Training**

Developers, enterprises, or researchers can publish datasets to Crust for Amadeus agents to use during training or evaluation.

#### **Cross-Chain User Access**

Users from the Polkadot ecosystem can authenticate and interact with Amadeus using SubWallet or Talisman, enabling broader participation.

#### **Durable Storage for uPoW Outputs**

Training outputs and inference artifacts produced under Useful Proof of Work are stored securely and verifiably for downstream consumption.

## **Current Status**

Integration is under active exploration, with technical design in progress and coordinated milestones across both teams.&#x20;