Fix ONNX Runtime CUDA fallback by preloading shared library dependencies

audio_separator/separator/architectures/mdx_separator.py

@@ -120,6 +120,18 @@ def load_model(self):
                 ort_session_options.log_severity_level = 0
 
             ort_inference_session = ort.InferenceSession(self.model_path, providers=self.onnx_execution_provider, sess_options=ort_session_options)
+            session_providers = ort_inference_session.get_providers()
+
+            requested_provider = self.onnx_execution_provider[0] if self.onnx_execution_provider else None
+            if requested_provider and requested_provider not in session_providers:
+                self.logger.warning(
+                    f"ONNX Runtime could not activate requested provider {requested_provider}; "
+                    f"session is using {session_providers}. This usually means required CUDA/cuDNN "
+                    f"runtime libraries are not visible to the dynamic loader."
+                )
+            else:
+                self.logger.debug(f"ONNX Runtime session providers: {session_providers}")
+
             self.model_run = lambda spek: ort_inference_session.run(None, {"input": spek.cpu().numpy()})[0]
             self.logger.debug("Model loaded successfully using ONNXruntime inferencing session.")
         else:

audio_separator/separator/separator.py

@@ -321,8 +321,26 @@ def setup_accelerated_inferencing_device(self):
         system_info = self.get_system_info()
         self.check_ffmpeg_installed()
         self.log_onnxruntime_packages()
+        self.preload_onnxruntime_dependencies()
         self.setup_torch_device(system_info)
 
+    def preload_onnxruntime_dependencies(self):
+        """
+        Preload ONNX Runtime shared library dependencies when supported by the installed package.
+
+        This helps pip-installed CUDA/cuDNN runtime wheels become visible to ONNX Runtime before
+        the first CUDAExecutionProvider session is created.
+        """
+        if not hasattr(ort, "preload_dlls"):
+            self.logger.debug("Installed ONNX Runtime does not provide preload_dlls(); skipping dependency preload.")
+            return
+
+        try:
+            ort.preload_dlls()
+            self.logger.debug("Preloaded ONNX Runtime shared library dependencies.")
+        except Exception as exc:
+            self.logger.warning(f"Unable to preload ONNX Runtime shared library dependencies: {exc}")
+
     def get_system_info(self):
         """
         This method logs the system information, including the operating system, CPU archutecture and Python version

audio_separator/utils/cli.py

@@ -150,13 +150,13 @@ def main():
         log_level = getattr(logging, args.log_level.upper())
     logger.setLevel(log_level)
 
-    from audio_separator.separator import Separator
-
     if args.env_info:
+        from audio_separator.separator import Separator
         separator = Separator()
         sys.exit(0)
 
     if args.list_models:
+        from audio_separator.separator import Separator
         separator = Separator(info_only=True)
 
         if args.list_format == "json":
@@ -190,6 +190,7 @@ def main():
         sys.exit(0)
 
     if args.list_presets:
+        from audio_separator.separator import Separator
         separator = Separator(info_only=True)
         presets = separator.list_ensemble_presets()
 
@@ -217,6 +218,7 @@ def main():
         sys.exit(0)
 
     if args.download_model_only:
+        from audio_separator.separator import Separator
         models_to_download = [args.model_filename] + (args.extra_models or [])
         separator = Separator(log_formatter=log_formatter, log_level=log_level, model_file_dir=args.model_file_dir)
         for model in models_to_download:
@@ -233,6 +235,8 @@ def main():
 
     logger.info(f"Separator version {package_version} beginning with input path(s): {', '.join(audio_files)}")
 
+    from audio_separator.separator import Separator
+
     separator = Separator(
         log_formatter=log_formatter,
         log_level=log_level,

tests/unit/test_cli.py

@@ -60,8 +60,15 @@ def test_cli_version_subprocess():
 
 # Test the CLI with no arguments
 def test_cli_no_args(capsys):
-    # Skip subprocess CLI tests - require proper CLI installation  
-    pytest.skip("CLI subprocess tests require proper installation")
+    test_args = ["cli.py"]
+
+    with patch("sys.argv", test_args), patch.dict("sys.modules", {"audio_separator.separator": None}):
+        with pytest.raises(SystemExit) as exc_info:
+            main()
+
+    assert exc_info.value.code == 1
+    captured = capsys.readouterr()
+    assert "Separate audio file into different stems." in captured.out
 
 
 # Test with multiple filename arguments

tests/unit/test_gpu_runtime_setup.py

@@ -0,0 +1,31 @@
+from unittest.mock import patch
+
+from audio_separator.separator.separator import Separator
+
+
+def test_setup_accelerated_inferencing_device_preloads_onnxruntime_dependencies():
+    separator = Separator(info_only=True)
+    system_info = object()
+
+    with patch.object(separator, "get_system_info", return_value=system_info), patch.object(separator, "check_ffmpeg_installed"), patch.object(
+        separator, "log_onnxruntime_packages"
+    ), patch("audio_separator.separator.separator.ort.preload_dlls", create=True) as mock_preload, patch.object(separator, "setup_torch_device") as mock_setup:
+        separator.setup_accelerated_inferencing_device()
+
+    mock_preload.assert_called_once_with()
+    mock_setup.assert_called_once_with(system_info)
+
+
+def test_setup_accelerated_inferencing_device_continues_when_preload_fails():
+    separator = Separator(info_only=True)
+    system_info = object()
+
+    with patch.object(separator, "get_system_info", return_value=system_info), patch.object(separator, "check_ffmpeg_installed"), patch.object(
+        separator, "log_onnxruntime_packages"
+    ), patch("audio_separator.separator.separator.ort.preload_dlls", side_effect=RuntimeError("boom"), create=True), patch.object(
+        separator, "setup_torch_device"
+    ) as mock_setup, patch.object(separator.logger, "warning") as mock_warning:
+        separator.setup_accelerated_inferencing_device()
+
+    mock_setup.assert_called_once_with(system_info)
+    mock_warning.assert_called_once()