AI Architecture & Systems
1
Microarchitecture
2
Kernel
3
Compiler
4
Architecture
5
System
6
Algorithms
Compiler
/
§3.13
The PyTorch Compilation Stack
3.13.1
TorchDynamo and FX Graph
#
3.13.2
AOTAutograd and Training-Graph Capture
#
3.13.3
TorchInductor, Triton, and CPU Codegen
#
3.13.4
Export, Dynamic Shape, and Custom Backends
#
3.13.5
Compilation Modes, Graph Breaks, and Performance Diagnosis
#
← Previous
3.12 Distributed Compilation
→ Next
3.14 JAX, XLA, and TPU Compilation