Phala Network
Phala Network is a hardware-secured compute platform for verifiable AI and confidential applications using Trusted Execution Environments. It provides GPU TEE support on NVIDIA H100, H200, and B200 chips alongside Intel and AMD CPU TEEs, enabling developers to deploy confidential AI models, containerized workloads via the dStack SDK, and confidential virtual machines with real-time cryptographic attestations. Phala targets privacy-preserving AI inference, enterprise data processing, and decentralized compute where verifiable execution and data confidentiality are required.
Tags
Overview
Phala Network is a decentralized confidential compute platform that uses Trusted Execution Environments to deliver verifiable AI and privacy-preserving applications. It combines CPU TEEs from Intel and AMD with GPU TEEs on NVIDIA H100, H200, and B200 hardware, allowing developers to run containerized workloads, confidential VMs, and AI inference with cryptographic attestations. The platform is SOC 2 Type I certified and HIPAA compliant.
Key Features
- GPU TEE — Confidential AI inference on NVIDIA H100, H200, and B200 accelerators
- dStack SDK — Deploy Docker and Kubernetes workloads inside TEEs
- Confidential VMs — Full virtual machines with hardware-backed isolation
- Real-time attestations — Cryptographic proofs of code and data integrity
- Pre-configured AI models — Ready-to-deploy confidential LLMs and inference endpoints