diff --git a/comfy/cli_args.py b/comfy/cli_args.py
index 4c654501..224c075f 100644
--- a/comfy/cli_args.py
+++ b/comfy/cli_args.py
@@ -84,7 +84,8 @@ parser.add_argument("--force-channels-last", action="store_true", help="Force ch
 
 parser.add_argument("--directml", type=int, nargs="?", metavar="DIRECTML_DEVICE", const=-1, help="Use torch-directml.")
 
-parser.add_argument("--disable-ipex-optimize", action="store_true", help="Disables ipex.optimize when loading models with Intel GPUs.")
+parser.add_argument("--oneapi-device-selector", type=str, default=None, metavar="SELECTOR_STRING", help="Sets the oneAPI device(s) this instance will use.")
+parser.add_argument("--disable-ipex-optimize", action="store_true", help="Disables ipex.optimize default when loading models with Intel's Extension for Pytorch.")
 
 class LatentPreviewMethod(enum.Enum):
     NoPreviews = "none"
diff --git a/comfy/model_management.py b/comfy/model_management.py
index d77ae8c0..244fa5c7 100644
--- a/comfy/model_management.py
+++ b/comfy/model_management.py
@@ -75,7 +75,7 @@ if args.directml is not None:
 try:
     import intel_extension_for_pytorch as ipex
     _ = torch.xpu.device_count()
-    xpu_available = torch.xpu.is_available()
+    xpu_available = xpu_available or torch.xpu.is_available()
 except:
     xpu_available = xpu_available or (hasattr(torch, "xpu") and torch.xpu.is_available())
 
@@ -219,12 +219,14 @@ if is_intel_xpu():
 if args.cpu_vae:
     VAE_DTYPES = [torch.float32]
 
-
 if ENABLE_PYTORCH_ATTENTION:
     torch.backends.cuda.enable_math_sdp(True)
     torch.backends.cuda.enable_flash_sdp(True)
     torch.backends.cuda.enable_mem_efficient_sdp(True)
 
+if int(torch_version[0]) == 2 and int(torch_version[2]) >= 5:
+    torch.backends.cuda.allow_fp16_bf16_reduction_math_sdp(True)
+
 if args.lowvram:
     set_vram_to = VRAMState.LOW_VRAM
     lowvram_available = True
diff --git a/main.py b/main.py
index b6504653..151b264c 100644
--- a/main.py
+++ b/main.py
@@ -114,6 +114,10 @@ if __name__ == "__main__":
         os.environ['HIP_VISIBLE_DEVICES'] = str(args.cuda_device)
         logging.info("Set cuda device to: {}".format(args.cuda_device))
 
+    if args.oneapi_device_selector is not None:
+        os.environ['ONEAPI_DEVICE_SELECTOR'] = args.oneapi_device_selector
+        logging.info("Set oneapi device selector to: {}".format(args.oneapi_device_selector))
+
     if args.deterministic:
         if 'CUBLAS_WORKSPACE_CONFIG' not in os.environ:
             os.environ['CUBLAS_WORKSPACE_CONFIG'] = ":4096:8"