[Bug]: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR

### Project Version 3.6.0 ### Platform and OS Version Windows 11 25H2 ### Affected Devices Wiindows PC ### Existing Issues _No response_ ### What happened? #1005 Using the method you suggested in the above issue(Thank you!), I was able to get the WebUI to launch, but an error occurs when I try to run inference. ``` An error occurred during audio conversion: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)` Traceback (most recent call last): File "C:\ApplioV3.6.0\rvc\infer\infer.py", line 291, in convert_audio audio_opt = self.vc.pipeline( ^^^^^^^^^^^^^^^^^ File "C:\ApplioV3.6.0\rvc\infer\pipeline.py", line 462, in pipeline pitch, pitchf = self.get_f0( ^^^^^^^^^^^^ File "C:\ApplioV3.6.0\rvc\infer\pipeline.py", line 239, in get_f0 f0 = model.get_f0(x, filter_radius=0.03) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ApplioV3.6.0\rvc\lib\predictors\f0.py", line 21, in get_f0 f0 = self.model.infer_from_audio(x, thred=filter_radius) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ApplioV3.6.0\rvc\lib\predictors\RMVPE.py", line 506, in infer_from_audio mel = self.mel_extractor(audio, center=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ApplioV3.6.0\env\Lib\site-packages\torch\nn\modules\module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ApplioV3.6.0\env\Lib\site-packages\torch\nn\modules\module.py", line 1541, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ApplioV3.6.0\rvc\lib\predictors\RMVPE.py", line 415, in forward mel_output = torch.matmul(self.mel_basis, magnitude) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)` ``` ### Steps to reproduce 1. Install HIP SDK 6.2.4 2. Run: ``` env\python -m pip uninstall torch torchvision torchaudio -y env\python -m pip install torch torchvision torchaudio --upgrade --index-url https://download.pytorch.org/whl/cu118 patch-zluda-hip62.bat run-applio-amd.bat ``` 3. inference ### Expected behavior The audio is processed correctly. ### Attachments _No response_ ### Screenshots or Videos _No response_ ### Additional Information GPU: Radeon RX 9060 XT AMD Software: Adrenalin Edition 25.9.2