AI Architecture & Systems

MLIR, TVM, and Deployment Compilation Stacks

3.15.1Multi-Level IR and Extensible Dialects in MLIR#

3.15.2TVM TensorIR, Relax, and Autotuning#

3.15.3IREE and Heterogeneous Runtimes#

3.15.4ONNX, ONNX Runtime, and Graph Interchange#

3.15.5TensorRT and Deployment Graph Optimization#

3.15.6Quantized Graph Conversion, Operator Coverage, and Backend Compatibility#