deployments/azure/README.md

Azure AI Foundry Deployment

Deploy SAM3 to Azure AI Foundry (pending GPU quota).

Quick Deploy

./deployments/azure/deploy.sh

This will build and push the image to Azure Container Registry.

Configuration

• Registry: sam3acr.azurecr.io
• Image: sam3-foundry:latest
• Endpoint: sam3-foundry (to be created)
• Resource Group: productionline-test
• Instance Type: Standard_NC6s_v3 (Tesla V100) or higher

Status

⏳ Pending GPU Quota Approval

Once GPU quota is approved, create the endpoint:

Create Endpoint (Azure Portal)

1. Navigate to Azure AI Foundry workspace
2. Go to Endpoints → Real-time endpoints
3. Click Create
4. Select Custom container
5. Image: sam3acr.azurecr.io/sam3-foundry:latest
6. Instance type: Standard_NC6s_v3 or higher
7. Deploy

Create Endpoint (Azure CLI)

# Create endpoint
az ml online-endpoint create \
  --name sam3-foundry \
  --resource-group productionline-test \
  --workspace-name <your-workspace>

# Create deployment
az ml online-deployment create \
  --name sam3-foundry-deployment \
  --endpoint sam3-foundry \
  --model-uri sam3acr.azurecr.io/sam3-foundry:latest \
  --instance-type Standard_NC6s_v3 \
  --instance-count 1

Testing

Once deployed, update the endpoint URL in the test script and run:

python3 scripts/test/test_api.py

For More Information

See docs/DEPLOYMENT.md for complete Azure AI Foundry deployment guide.