Deploy your AI models in a secure, isolated environment with hardware-level protection. Phala Cloud’s Confidential AI Models service runs your models inside TEE on GPU hardware, ensuring your data and model weights stay private.

​
Deploy Your Model

Navigate to the Confidential AI Models tab in your Phala Cloud dashboard. You’ll see available model templates ready for deployment.
Confidential AI Models
Don’t see your preferred model? We can add custom models—just reach out through Support.
Pick your model and hit Deploy to get started.

​
Configure Your Deployment

You’ll need to fill in a few details to customize your setup:
  1. Service Name: Give your deployment a memorable name (e.g., “qwen3-8b-d7a7s”)
  2. Model Template: Select your preferred model (e.g., “qwen3”)
  3. Node & Image: Keep the default settings unless you have specific requirements
  4. Resource Plan: Choose the instance type that matches your workload needs
  5. Scheduled Destroy: Set an auto-destruction date if needed, or leave empty for manual control
  6. Review: Check your configuration in the deployment summary
Deploy LLM Model
Review your settings and click Deploy to launch.
Our team will contact you directly to complete the rest of the configuration and guide you through the final steps.