Model Installer

Component Category

Model lifecycle and model delivery

Component Description

Model Installer is the CoreAI service that connects model sources, model artifacts, model registry records, and the Runtime inference layer. It handles the operational work required to move models from import and registration into usable inference deployments.

Why It Is Used

It standardizes how models move from tracked artifacts into deployable inference assets, which helps teams manage model onboarding more consistently. It is especially useful when models need to be pulled from Hugging Face or S3-compatible storage, registered in MLflow, and then exposed to Runtime inference services.

Learn More

• CoreAI Reference Architecture
• How CoreAI Builds On Runtime

Typical Flow

1. Import a model from Hugging Face or from S3-compatible storage.
2. Upload or register the model artifacts in MLflow.
3. Optionally connect those artifacts to an inference registration flow.
4. Register the model into the Runtime inference environment, especially KubeAI.
5. Make the model available to higher-level CoreAI services.

Practical Notes

• It supports both REST and gRPC APIs.
• It can process long-running imports asynchronously and track task progress.
• It can register models already stored in S3 directly in MLflow without re-uploading them.
• It is the operational bridge between model lifecycle management and actual inference availability.

Interacts With

• MLflow for model registry and lifecycle information
• MinIO for model artifacts and related files
• LiteLLM for internal proxy registration and shared model access patterns
• Keycloak for authenticated access
• OpenTelemetry for service telemetry
• KubeAI, vLLM, and other Runtime inference services where installed models are consumed