

The TrustEE Abstraction Stack
Deploy verifiable onchain AI agents without bare-metal overhead. TrustEE translates complex hardware enclaves into clean, developer-ready cryptographic proofs.




Multi-Enclave Architecture
Unified Enclave Interface
Our middleware abstracts Intel SGX, AMD SEV, and AWS Nitro into a single API. Build your model once and deploy across diverse secure hardware environments.
State Attestation Contracts
Onchain verification contracts translate hardware-level attestation directly into smart contract states, enabling trustless execution verification on any EVM chain.
Three Steps to Verifiable AI
Model Packaging
Attestation Generation
Onchain Verification
Compile your multimodal language model or agent logic into a secure, hardware-isolated WebAssembly or Docker enclave container.
TrustEE generates zero-latency cryptographic proofs directly at the hardware execution level during every model run.
Submit the generated proof to our lightweight verification smart contracts to trigger automated trustless state transitions.


System Architecture FAQ
Have deeper integration questions? Our developer channel is active 24/7.
What is the performance overhead?
Which hardware is supported?
How is the proof verified onchain?
By optimizing the TrustEE abstraction layer, we reduce TEE execution latency to under 5 milliseconds, making it viable for high-frequency DeFi agents.
We support Intel SGX, AMD SEV-SNP, and AWS Nitro Enclaves. The middleware automatically selects the optimal environment based on workload requirements.
TrustEE translates hardware attestation reports into succinct cryptographic proofs that are verified via gas-optimized Solidity contracts.
