George S

@BDHU - I see, yes there is a compilation argument, `truncate_long_and_double`, which will handle the casting of int64 to int32 directly. You would just specify `"truncate_long_and_double": True` as a kwarg....

# Analysis Findings - Inference results [FP16] for both Llama and SDXL models in Torch-TensorRT's `torch.compile` backend have accuracy discrepancies relative to the Torch counterpart model - Specifically, the inference...

### Update We have further narrowed the matmul cases for easier example-reproducing ### Next Steps - See if issues persist when using FP32 precision, again narrow down the cases to...

Add-on to #2324

Hi - it looks like the model code itself is setting attributes of the `nn.Module`, which causes issues for some of our TorchScript lowering passes. If this is a detection-style...

Thanks for the follow-up. The `SetAttr` issue appears to be thrown here, in the `torch::jit::LowerGraph` function from PyTorch. https://github.com/pytorch/TensorRT/blob/20264a3c03065fce089cf284b5e172c50cc3bc14/core/lowering/lowering.cpp#L181 From the model code, it appears some attribute `self.W` integer parameter...

Thank you for testing this out - it looks like an issue is encountered when tracing the model in the Dynamo path. Could you also check `ir="torch_compile"` and see if...

Hi - this appears to be an issue during scripting of the model with `torch.jit.script(module)`. Could you try `ts_model = torch.jit.script(model)` to see if the model can be scripted to...

# Overview of Findings - Torch-TRT wheels now building successfully in CI within PR #2538 using TRT 9 - Converters failing due to data type changes/mismatches (seem to be minor...

Hi - thank you for the report. I noticed these results are using the `23.08-py3` container, and there have been many upgrades and changes to Torch-TensorRT since the version which...