Skip to main content

​
Custom models and bring-your-own-key

You can register your own model deployments and API keys with Opper. Custom models appear alongside built-in models and get the same observability, routing, and fallback features.

​
Register a custom model

curl -X POST https://api.opper.ai/v2/custom-models \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${OPPER_API_KEY}" \
  -d '{
    "name": "example/my-gpt4",
    "provider": "azure",
    "model_id": "gpt4-production",
    "api_key": "my-api-key-here",
    "params": {
      "api_base": "https://my-gpt4-deployment.openai.azure.com/",
      "api_version": "2024-06-01"
    }
  }'
Custom models are scoped to your project. The model name prefix (e.g., example/) serves as a namespace.

​
Model aliases

Create aliases to abstract model names, enabling zero-downtime model swaps: 
curl -X POST https://api.opper.ai/v2/model-aliases \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${OPPER_API_KEY}" \
  -d '{
    "name": "production/main",
    "model_id": "anthropic/claude-sonnet-4"
  }'
Use production/main in your code, then update the alias target to switch models without changing any integration code.