Torch-FL#
Torch-FL (torch_fl) is a PyTorch device plugin for the FlagOS software stack. It exposes a single flagos device that routes operators among reusable native kernels, portable compiler kernels, vendor-native implementations, and explicit CPU fallback, so the same PyTorch program runs across accelerators without workload changes.
What Torch-FL is, its design principles, capabilities, and component architecture.
Platform selection, build requirements, environment setup, verification, and test markers.
ProcessGroupFlagOS, FlagCX, vendor fallbacks, DDP and DataParallel support.
torch.profiler integration, vendor tracers, correlation ids, and parity with torch.cuda.
Inductor integration for the flagos device, FlagTree compilation, and platform notes.
Per-platform capability validation, environment variables, and operation-routing configuration.