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	1213	January 9, 2023
Building from GIT error: PyTorch needs CUDNN-8.5 or above but found 8.2.4 torch.compile	4	17	June 17, 2025
Preserving bits like conj, neg during functionalization FX (Functional Transformations)	0	8	June 12, 2025
How to obtain input variable of each operand of a fx ir torch.compile	1	14	June 10, 2025
What options do I have when torch.compile with TF32 matmuls causes accuracy issues? torch.compile	0	12	June 9, 2025
Torch compile: optimizer.step Generates Excessive Warning Messages torch.compile	11	349	June 5, 2025
Torch.compile usage for generative models torch.compile	5	49	May 28, 2025
How to get positional order of inputs and outputs for a graph that was exported via torch.export? torch.compile	1	111	May 26, 2025
PyTorch Error: Compiler: cl is not found torch._inductor	2	228	May 21, 2025
How to customize leaf node in dynamo trace? torch._dynamo	2	570	May 21, 2025
Torch.compile and Parameter replacement in a module torch.compile	0	15	May 20, 2025
How to free backward input args when I use iree as the compilation backend? torch.compile	0	13	May 20, 2025
In inductor, can we unify the way to generate code between using extern_kernels and torch.ops? torch._inductor	1	79	May 19, 2025
[PyTorch 2.7.0] Custom privateuse1 FakeTensor add + float scalar triggers AttributeError: 'float' object has no attribute 'item_memo' torch._dynamo	0	39	May 19, 2025
Torch not compiled with CUDA enabled for ComfyUI torch.compile	1	181	May 13, 2025
Where the grades from backward graph will be copied to param.grade in cuda graphs torch.compile	0	38	May 7, 2025
Recompilations leading to timeout torch.compile	0	49	May 1, 2025
torch._dynamo.exc.Unsupported: call_method NNModuleVariable() _sa_block [TensorVariable(), LazyVariableTracker(), LazyVariableTracker()] {} torch._dynamo	1	480	April 29, 2025
How to use triton kernels with pytorch? torch.compile	1	85	April 29, 2025
What is ComputedBuffer? torch._inductor	1	267	April 24, 2025
Compiled matmul is slower than vanilla matmul torch.compile	1	82	April 24, 2025
FP8 `torch.empty` doesn't work under `inductor` of pytorch 2.4.1 torch.compile	2	74	April 24, 2025
Device "meta" and device "cuda:0" error torch.compile	0	68	April 23, 2025
Resetting cache in benchmark torch._dynamo	4	1629	April 23, 2025
Torch.compile cpp backend not fusing OuterLoopFusedSchedulerNode properly torch._inductor	2	67	April 22, 2025
Higher Peak Memory on First Run with torch.compile and Inductor Backend torch.compile	0	25	April 21, 2025
How to prevent dispatching `torch._C._nn.upsample_bilinear2d` to vectorize implementation `torch.ops.aten.upsample_bilinear2d.vec` torch.compile	0	16	April 21, 2025
Why adaptive_avg_pool1d became adaptive_avg_pool2d in make_fx? FX (Functional Transformations)	1	55	April 21, 2025
C++/Cuda / aot.compile and cuda graph torch.compile	1	507	April 18, 2025
How does torch dynamo set custom eval frame callback? torch._dynamo	1	28	April 16, 2025