[Python] Is torch_mlir.compile currently part of the official PyTorch or torch-mlir API?

I’ve seen references to torch_mlir.compile from several AI tools especially from ChatGPT, DeepSeek, etc. I'm also attaching a sample code based on the scenario.

import torch
import torch_mlir

# Define a simple PyTorch model
class SimpleModel(torch.nn.Module):
def __init__(self):
super(SimpleModel, self).__init__()
self.linear = torch.nn.Linear(3, 2)

def forward(self, x):
return self.linear(x)

# Create an instance of the model
model = SimpleModel()
model.eval() # Set model to evaluation mode

# Create example input tensor
example_input = torch.randn(1, 3)

# Compile the model using Torch-MLIR
mlir_module = torch_mlir.compile(model, (example_input,), output_type="torch")

# Print the MLIR output
print(mlir_module)


Is torch_mlir.compile an officially supported function in the latest versions of PyTorch or torch-mlir, or is it outdated or renamed?

I’d appreciate a clear answer based on the current state of the API.

Continue reading...