CLI Examples

This page provides practical examples of common PaiTIENT Secure Model CLI usage scenarios.

Basic Commands

Authenticate with the CLI

# Interactive login
secure-model login

# Or set environment variables
export PAITIENT_API_KEY="your-api-key"
export PAITIENT_CLIENT_ID="your-client-id"

List Available Models

# List all available models
secure-model list models

# Filter models by keyword
secure-model list models --filter "clinical"

# Output in JSON format
secure-model list models --output json

Model Deployment

Basic Model Deployment

# Deploy a model with default settings
secure-model deploy model --name ZimaBlueAI/HuatuoGPT-o1-8B --deployment-name clinical-assistant

# Show deployment progress
secure-model status --deployment-id dep_12345abcde

# List all deployments
secure-model list deployments

Advanced Deployment Options

# Deploy with specific compute resources
secure-model deploy model \
  --name ZimaBlueAI/HuatuoGPT-o1-8B \
  --deployment-name clinical-assistant \
  --compute-type gpu \
  --instance-type g4dn.xlarge \
  --min-replicas 1 \
  --max-replicas 3 \
  --auto-scaling true

# Deploy with tags
secure-model deploy model \
  --name ZimaBlueAI/HuatuoGPT-o1-8B \
  --deployment-name clinical-assistant \
  --tags environment=production,department=clinical-research

# Deploy with timeout
secure-model deploy model \
  --name ZimaBlueAI/HuatuoGPT-o1-8B \
  --deployment-name clinical-assistant \
  --timeout 30m

Text Generation

Basic Text Generation

# Generate text with a simple prompt
secure-model generate \
  --deployment-id dep_12345abcde \
  --prompt "What are the potential side effects of metformin?"

# Pipe input from a file
cat prompt.txt | secure-model generate --deployment-id dep_12345abcde

# Save output to a file
secure-model generate \
  --deployment-id dep_12345abcde \
  --prompt "What are the potential side effects of metformin?" \
  > response.txt

Advanced Generation Options

# Generate with specific parameters
secure-model generate \
  --deployment-id dep_12345abcde \
  --prompt "What are the potential side effects of metformin?" \
  --max-tokens 500 \
  --temperature 0.7 \
  --top-p 0.95 \
  --stop "\n\n"

# Stream response in real-time
secure-model generate \
  --deployment-id dep_12345abcde \
  --prompt "What are the potential side effects of metformin?" \
  --stream

# Interactive chat mode
secure-model chat --deployment-id dep_12345abcde

Deployment Management

Check Deployment Status

# Get detailed deployment status
secure-model status --deployment-id dep_12345abcde

# Get deployment metrics
secure-model metrics --deployment-id dep_12345abcde

# Watch deployment status in real-time
secure-model status --deployment-id dep_12345abcde --watch

Update Deployment

# Scale replicas
secure-model update deployment \
  --deployment-id dep_12345abcde \
  --min-replicas 2 \
  --max-replicas 5

# Update tags
secure-model update deployment \
  --deployment-id dep_12345abcde \
  --tags environment=staging,department=clinical-research

# Enable auto-scaling
secure-model update deployment \
  --deployment-id dep_12345abcde \
  --auto-scaling true

Delete Deployment

# Delete a deployment
secure-model delete deployment --deployment-id dep_12345abcde

# Delete with confirmation bypass
secure-model delete deployment --deployment-id dep_12345abcde --force

Working with Secure Batches

Create Batch Job

# Create a batch job with inputs from a file
secure-model create batch \
  --deployment-id dep_12345abcde \
  --input-file inputs.jsonl \
  --output-file outputs.jsonl

# Create a batch job with specific parameters
secure-model create batch \
  --deployment-id dep_12345abcde \
  --input-file inputs.jsonl \
  --output-file outputs.jsonl \
  --max-tokens 500 \
  --temperature 0.7

Monitor Batch Job

# Check batch job status
secure-model status --batch-id batch_12345abcde

# Watch batch job progress
secure-model status --batch-id batch_12345abcde --watch

Fine-tuning Models

Prepare Data for Fine-tuning

# Validate and prepare training data
secure-model prepare fine-tuning \
  --input-file training_data.jsonl \
  --output-file prepared_data.jsonl

Start Fine-tuning

# Fine-tune a model
secure-model fine-tune \
  --model ZimaBlueAI/HuatuoGPT-o1-8B \
  --training-file prepared_data.jsonl \
  --validation-file validation_data.jsonl \
  --epochs 3

# Fine-tune with advanced parameters
secure-model fine-tune \
  --model ZimaBlueAI/HuatuoGPT-o1-8B \
  --training-file prepared_data.jsonl \
  --validation-file validation_data.jsonl \
  --epochs 3 \
  --learning-rate 5e-5 \
  --batch-size 8

Monitor Fine-tuning

# Check fine-tuning status
secure-model status --fine-tuning-id ft_12345abcde

# Check fine-tuning metrics
secure-model metrics --fine-tuning-id ft_12345abcde

Configuration and Setup

Configure CLI

# Interactive configuration
secure-model configure

# Create a new profile
secure-model configure --profile production

Install Auto-completion

# For bash
secure-model completion bash > ~/.secure-model-completion.bash
echo 'source ~/.secure-model-completion.bash' >> ~/.bashrc

# For zsh
secure-model completion zsh > ~/.secure-model-completion.zsh
echo 'source ~/.secure-model-completion.zsh' >> ~/.zshrc

Advanced Usage

Using JSON Input

# Deploy using JSON configuration file
secure-model deploy model --config deployment-config.json

# Generate using JSON request file
secure-model generate --request-file generate-request.json

Using the CLI in Scripts

# Use jq to parse JSON output
DEPLOYMENT_ID=$(secure-model deploy model \
  --name ZimaBlueAI/HuatuoGPT-o1-8B \
  --deployment-name clinical-assistant \
  --output json | jq -r '.deployment_id')

echo "Deployment ID: $DEPLOYMENT_ID"

# Wait for deployment to be ready
while true; do
  STATUS=$(secure-model status --deployment-id $DEPLOYMENT_ID --output json | jq -r '.status')
  echo "Current status: $STATUS"
  
  if [ "$STATUS" = "running" ]; then
    break
  elif [ "$STATUS" = "failed" ]; then
    echo "Deployment failed"
    exit 1
  fi
  
  sleep 30
done

# Generate text once deployment is ready
secure-model generate \
  --deployment-id $DEPLOYMENT_ID \
  --prompt "What are the potential side effects of metformin?"

Next Steps

• Learn about CLI Commands
• Understand CLI Configuration
• Review Authentication
• Check Troubleshooting