Search results

torch.prims.convert_element_type to linalg bf16 to f16 fail

Status: Open.

#3962 In llvm/torch-mlir;

· AmosLewis opened on Jan 16, 2025

[cmake] Torch-MLIR doesn't build with CMake setting 'BUILD_SHARED_LIBS=ON'

Status: Open.

#3961 In llvm/torch-mlir;

· christopherbate opened on Jan 15, 2025

[QUERY] How to stop torch-mlir from decomposing certain ops when using torch dynamo

Status: Open.

#3953 In llvm/torch-mlir;

· Abhishek-TyRnT opened on Jan 10, 2025

Lowering program that mutates its inputs - how to lower torch.copy.to_vtensor and torch.overwrite.tensor.contents ops

Status: Open.

#3944 In llvm/torch-mlir;

· dixinzhou opened on Jan 7, 2025

attention model failed to convert, "error: unknown: unsupported by backend contract: module initializers"

Status: Open.

#3915 In llvm/torch-mlir;

· heshuju opened on Dec 10, 2024

Fix LTC build error in CI

Status: Open.

#3910 In llvm/torch-mlir;

· vivekkhandelwal1 opened on Dec 9, 2024

Failed to compile onnx.Resize with dynamic shapes

Status: Open.

#3903 In llvm/torch-mlir;

· jinchen62 opened on Dec 4, 2024

torch.aten._scaled_dot_product_efficient_attention：Failed to legalize operation 'torch.operator' that was explicitly marked illegal

Status: Open.

#3893 In llvm/torch-mlir;

· bilu-1 opened on Nov 28, 2024

Assertion when lowering from Torch IR for AvgPool2d when kernel is a tuple of a single int

Status: Open.

#3885 In llvm/torch-mlir;

· sahas3 opened on Nov 21, 2024

Llama 3.1 8B F16 Non-Decomposed prefill fails to compile

Status: Open.

#3884 In llvm/torch-mlir;

· aviator19941 opened on Nov 19, 2024

Failing tests for check-torch-mlir

Status: Open.

#3881 In llvm/torch-mlir;

· mdehling opened on Nov 19, 2024

failed to legalize operation 'torch.aten.cumsum'

Status: Open.

#3866 In llvm/torch-mlir;

· DavidGinten opened on Nov 12, 2024