Model Serving · Interface

Agent Packaging Interface

InterfaceModel ServingModelsarc:AgentPackagingInterface

A standard load-context and predict contract wrapping an agent so any compatible serving platform can load its artifacts and invoke it without deployment-specific code.

Responsibility. Decouples a packaged agent from the platform that serves it.

Also known as: PythonModel wrapper, Model-serving predict contract

Direct neighbourhood (hover for relationship types)

Relationships

is exposed by structural

Agent Controller abstract Ch4.2

is invoked by dependency

Inference Server Ch4.2

Design guidance

SHOULD return structured outputs including success status, response and execution metadata (tool calls, reasoning steps).

Classification

Patterns: AdapterStandardised model packaging
Technologies: MLflow PythonModelMLflow model serverSeldonAmazon SageMaker
Quality attributes: Flexibility (ISO/IEC 25010)Compatibility (ISO/IEC 25010)
Risks mitigated: Platform lock-in of agent services

Sources

Ch4.2: T. Nguyen, "Deployment and Scaling," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 4.2. ISBN: 9798244538229.