News
DealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory鈥檚 NVIDIA & AMD Enterprise ExpansionDealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory鈥檚 NVIDIA & AMD Enterprise Expansion
EveryNeocloud
Phala Cloud

Phala Cloud

Verifiable private AI compute on GPU TEEs

Phala Cloud offers confidential computing for AI workloads using hardware-backed Trusted Execution Environments. It allows developers to run private inference and GPU jobs with cryptographic proof of execution, ensuring data privacy without sacrificing verifiability.

Share on WhatsApp

  • Runs H100, H200, and B300 GPUs in TEEs
  • Provides cryptographic attestation for runtime verification
  • Supports Docker Compose for easy deployment
  • Claims 5-10% performance overhead for confidential workloads
  • SOC 2 Type I certified and HIPAA compliant

What it is

Phala Cloud provides a platform for running AI workloads in confidential environments. It uses Trusted Execution Environments (TEEs) to keep data and models private during processing. The service supports Docker-based deployments and offers verifiable runtime proofs. Users can run agents, large language models, and GPU-intensive tasks while maintaining strict privacy controls. The system generates cryptographic attestations that confirm the integrity of the code and hardware used.

How it came about

The project emerged from Phala Network, focusing on verifiable private compute. The team expanded from CPU enclaves to include GPU capacity for AI applications. They developed dstack as part of their tooling ecosystem. The company positions itself as building the proof layer necessary for private AI adoption. It targets developers and enterprises needing secure execution without trusting cloud providers blindly.

The compute

The platform supports Nvidia H100, H200, and B300 GPUs within TEEs. It aims for near-native performance with an estimated five to ten per cent overhead. Workloads run in confidential virtual machines with sealed keys and private memory. The system integrates with standard AI frameworks like PyTorch and TensorFlow. Attestations are signed by Intel or Nvidia to verify hardware and runtime state.

Customers and deals

The company claims enterprise clients use its service for sensitive trading data and healthcare applications. It states users have reduced compliance costs and accelerated drug discovery. Fortune 500 companies are cited as customers requiring verifiable data protection. The platform supports multi-party collaboration on private datasets.

Its place in AI

Phala Cloud operates as a decentralized GPU and AI economy. It enables autonomous agents with verifiable execution records on-chain. The service aligns with regulatory standards such as GDPR and HIPAA. It holds SOC 2 Type I certification and is pursuing ISO 27001. The model emphasizes public proof for private AI operations.

Questions people ask about Phala Cloud

What hardware does Phala Cloud support?

Phala Cloud offers Nvidia H100, H200, and B300 GPUs. These chips run inside hardware-backed Trusted Execution Environments to ensure data privacy and verifiable execution.

How does Phala verify workload integrity?

The platform generates cryptographic attestations signed by Intel or Nvidia. These proofs confirm the code, image, and hardware state, allowing users to verify execution without seeing private data.

Is Phala Cloud compatible with existing AI tools?

Yes, it supports Docker services and popular frameworks like PyTorch, TensorFlow, and Hugging Face. Developers can deploy existing Docker Compose workloads with minimal changes to the configuration.

Good reading and sources

    Where this page comes from

    There is no English Wikipedia article on Phala Cloud yet, so the facts on this page come from the sources listed above: phala.network/about, phala.network/. Some are the subject's own pages, so its claims about itself are reported as its claims. EveryNeocloud wrote the text in its own words, with AI assistance, and compares it with the sources to make sure no passage is copied.

    It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.

    Photo credits on this page

    The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

    The weekly digest

    Every Monday: the provider of the day, new pages on EveryNeocloud, and the conferences and anniversaries of the week ahead.