torch.compile

torch._dynamo PyTorch’s compiler frontend which can output subgraphs when programs aren’t traceable. FX (Functional Transformations) FX is a toolkit for developers to use to transform nn.Module instances. FX consists of three main components: a symbolic tracer, an intermediate representation , and Python code generation . torch._inductor The default compiler in torch.compile, builds loops schedules and codegens triton or C++ kernels.

Topic	Replies	Views	Activity
About the torch.compile category torch.compile	0	1261	January 9, 2023
Torch.compile + flex_attention with custom score_mod fails (FlexibleLayout / NoValidChoicesError) torch.compile	0	8	September 19, 2025
Are there any related works on "megakernel"? torch.compile	0	20	September 17, 2025
Coordinate_descent_tuning errors out with torch.AcceleratorError: CUDA error: invalid argument torch.compile	10	54	September 15, 2025
In Dynamo+AOTAutograd, why run Faketensor through the code multiple times? torch._dynamo	0	11	September 15, 2025
Better understanding why AOTAutograd decomposes `fused_rms_norm_backward` for CUDA, but not for Meta tensors torch._dynamo	0	12	September 15, 2025
Compile: `DataLoader.__setattr__` should not be traced torch.compile	2	24	September 13, 2025
Torch.compile error on backward torch.compile	0	14	September 12, 2025
Do I need to rerun torch.compile between training and inference? torch.compile	1	28	September 9, 2025
Autofunctionalize v2 / Reinplace FX (Functional Transformations)	0	13	September 9, 2025
Torch.compile does not work after first graph break torch.compile	0	21	September 5, 2025
Module 'torch.library' has no attribute 'custom_op' torch.compile	5	661	September 5, 2025
How to understand the files generated by TORCH_COMPILE_DEBUG FX (Functional Transformations)	0	14	September 4, 2025
Dynamo logs generated by torch.distributed torch.compile	5	59	September 3, 2025
Choice of torch.compile vs. triton torch.compile	7	7908	August 29, 2025
Reducing compiled call overhead torch.compile	0	27	August 29, 2025
PyTorch Error: Compiler: cl is not found torch._inductor	3	1062	August 27, 2025
How can we reliably figured out what the parameters of a generated PTX kernel are with torch.compile()? torch.compile	0	21	August 25, 2025
C++/Cuda / aot.compile and cuda graph torch.compile	2	549	August 23, 2025
Training with flex attention is extremely slow due to torch.compile settings torch.compile	0	68	August 22, 2025
Is it possible that node's name can change during torch.compile? torch.compile	0	15	August 22, 2025
Torch.compile Numpy code throws mean() arguments error torch.compile	0	19	August 21, 2025
Using RNG Generator in torch.compile torch.compile	2	52	August 20, 2025
How to turn off inlining / force materialization in TorchInductor during torch.compile? torch._inductor	0	19	August 16, 2025
Some questions about torch.compile torch.compile	0	26	August 15, 2025
`torch.compile` (w/ Torch Inductor) benchmarks/models for Multi GPU torch._inductor	0	43	August 7, 2025
CUDA Graph Error with Residual Connections in `torch.compile` (RuntimeError: accessing tensor output of CUDAGraphs) torch.compile	0	62	August 1, 2025
The CUDA kernel produces different results when running in CUDA Graph mode compared to non-CUDA Graph mode torch.compile	0	20	July 30, 2025
Skip dynamo when using 'cudagraph' backend torch.compile	0	36	July 28, 2025
Getting the fx graph of submodules, instead of 'call_module' nodes? torch.compile	1	747	July 24, 2025