PaiTIENT Secure Model Service

Introduction

The PaiTIENT Secure Model Service provides a HIPAA/SOC2 compliant platform for deploying and managing AI models with industry-leading security and privacy features. This guide will help you understand how to use both the Python and Node.js SDKs to securely deploy models, generate predictions, and manage your deployments.

Key Features

• Hybrid Encryption: Models are encrypted using AES-256-GCM with Kyber post-quantum encryption for keys
• HIPAA/SOC2 Compliance: All operations follow strict compliance guidelines for healthcare data
• Secure Deployment: Isolated AWS and Kubernetes environments for each client
• Client-Specific Keys: Each client has their own encryption keys
• LoRA Fine-tuning: Support for fine-tuning models on private data
• Complete Audit Trail: Comprehensive logging and monitoring

Supported Models

PaiTIENT Secure Model Service supports a variety of models, including:

• ZimaBlueAI/HuatuoGPT-o1-8B (Primary supported model)
• Other HuggingFace compatible models (Custom deployments available)

Where to Start

If you're new to PaiTIENT Secure Model Service, we recommend starting with the following guides:

1. Installation - Learn how to install the SDKs for Python and Node.js
2. Quick Start - Get up and running with basic model deployment
3. Python SDK Guide - Comprehensive guide for Python users
4. Node.js SDK Guide - Comprehensive guide for Node.js users

Authentication

All interactions with the PaiTIENT Secure Model Service require authentication. You will need:

• An API Key
• A Client ID
• An endpoint URL

These will be provided when you sign up for the service. Learn more in our authentication guide.

Example Usage (Python)

from paitient_secure_model import SecureModelClient

# Initialize the client
client = SecureModelClient(
    api_key="your-api-key",
    client_id="your-client-id",
    endpoint="https://api.paitient.ai"
)

# Deploy a model
deployment = client.deploy_model(
    model_name="ZimaBlueAI/HuatuoGPT-o1-8B",
    deployment_name="my-secure-model"
)

# Generate text
result = client.generate(
    deployment_id=deployment.id,
    prompt="Explain the importance of HIPAA compliance in healthcare AI",
    max_tokens=500
)

print(result.text)

Example Usage (Node.js)

const { SecureModelClient } = require('paitient-secure-model');

// Initialize the client
const client = new SecureModelClient({
  apiKey: "your-api-key",
  clientId: "your-client-id",
  endpoint: "https://api.paitient.ai"
});

// Deploy a model
async function deployAndUseModel() {
  const deployment = await client.deploy({
    modelName: "ZimaBlueAI/HuatuoGPT-o1-8B",
    deploymentName: "my-secure-model"
  });

  // Generate text
  const result = await client.generate({
    deploymentId: deployment.id,
    prompt: "Explain the importance of HIPAA compliance in healthcare AI",
    maxTokens: 500
  });

  console.log(result.text);
}

deployAndUseModel();