databricks-model-serving - Manage Databricks model serving endpoints
Manage Databricks serving endpoint lifecycles, traffic routing, model queries, schemas, operations, AI Gateway limits, and integrations.
Tags
Updated: 2026-10-02Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Create serving endpoints
- Update endpoint configurations
- Route endpoint traffic
- Swap model versions
- Query served models
- Stream endpoint responses
- Retrieve OpenAPI schemas
- Inspect endpoint operations
- Manage AI Gateway limits
- Discover foundation model endpoints
- Integrate Databricks Apps
- Stream off-platform clients
Inputs
- Endpoint names
- Endpoint configurations
- Model references
- Model versions
- Traffic routes
- Query payloads
- Databricks CLI profile
- Authentication credentials
- Client requests
Outputs
- Serving endpoints
- Endpoint state changes
- Model query responses
- Streaming responses
- OpenAPI schemas
- Logs
- Metrics
- Permission information
- Rate limit configurations
Requirements
- Databricks CLI >= v0.294.0
- Databricks authentication
- Databricks CLI profile
