Supports both standard PMML and ONNX models
Integrates PMML4S to score PMML models
Depends on ONNX Runtime to support ONNX models with CPU and GPU accelerators
Provides both REST and gPRC endpoints
AI-Serving
Serve PMML and ONNX models with both REST and gRPC endpoints
Description
AI-Serving is a flexible, high-performance inferencing system for machine learning and deep learning models, designed for production environments. AI-Serving provides out-of-the-box integration with PMML and ONNX models, but can be easily extended to serve other formats of models.