Merge 96e5287a72 into 025e6792ee

comfy/cli_args.py

@@ -91,7 +91,6 @@ parser.add_argument("--directml", type=int, nargs="?", metavar="DIRECTML_DEVICE"
 
 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("--supports-fp8-compute", action="store_true", help="ComfyUI will act like if the device supports fp8 compute.")
-parser.add_argument("--enable-triton-backend", action="store_true", help="ComfyUI will enable the use of Triton backend in comfy-kitchen. Is disabled at launch by default.")
 
 class LatentPreviewMethod(enum.Enum):
     NoPreviews = "none"

comfy/model_management.py

@@ -721,15 +721,13 @@ def load_models_gpu(models, memory_required=0, force_patch_weights=False, minimu
     else:
         minimum_memory_required = max(inference_memory, minimum_memory_required + extra_reserved_memory())
 
-    # Order-preserving dedup. A plain set() would randomize iteration order across runs
-    models_temp = {}
+    models_temp = set()
     for m in models:
-        models_temp[m] = None
+        models_temp.add(m)
         for mm in m.model_patches_models():
-            models_temp[mm] = None
+            models_temp.add(mm)
 
-    models = list(models_temp)
-    models.reverse()
+    models = models_temp
 
     models_to_load = []
 

comfy/model_prefetch.py

@@ -37,8 +37,7 @@ def prefetch_queue_pop(queue, device, module):
     consumed = queue.pop(0)
     if consumed is not None:
         offload_stream, prefetch_state = consumed
-        if offload_stream is not None:
-            offload_stream.wait_stream(comfy.model_management.current_stream(device))
+        offload_stream.wait_stream(comfy.model_management.current_stream(device))
         _, comfy_modules = prefetch_state
         if comfy_modules is not None:
             cleanup_prefetched_modules(comfy_modules)

comfy/ops.py

@@ -253,9 +253,6 @@ def resolve_cast_module_with_vbar(s, dtype, device, bias_dtype, compute_dtype, w
     if bias is not None:
         bias = post_cast(s, "bias", bias, bias_dtype, prefetch["resident"], update_weight)
 
-    if prefetch["signature"] is not None:
-        prefetch["resident"] = True
-
     return weight, bias
 
 

comfy/quant_ops.py

@@ -1,8 +1,6 @@
 import torch
 import logging
 
-from comfy.cli_args import args
-
 try:
     import comfy_kitchen as ck
     from comfy_kitchen.tensor import (
@@ -23,15 +21,7 @@ try:
             ck.registry.disable("cuda")
             logging.warning("WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations.")
 
-    if args.enable_triton_backend:
-        try:
-            import triton
-            logging.info("Found triton %s. Enabling comfy-kitchen triton backend.", triton.__version__)
-        except ImportError as e:
-            logging.error(f"Failed to import triton, Error: {e}, the comfy-kitchen triton backend will not be available.")
-            ck.registry.disable("triton")
-    else:
-        ck.registry.disable("triton")
+    ck.registry.disable("triton")
     for k, v in ck.list_backends().items():
         logging.info(f"Found comfy_kitchen backend {k}: {v}")
 except ImportError as e:

comfy/sampler_helpers.py

@@ -89,8 +89,7 @@ def get_additional_models(conds, dtype):
         gligen += get_models_from_cond(conds[k], "gligen")
         add_models += get_models_from_cond(conds[k], "additional_models")
 
-    # Order-preserving dedup. A plain set() would randomize iteration order across runs
-    control_nets = list(dict.fromkeys(cnets))
+    control_nets = set(cnets)
 
     inference_memory = 0
     control_models = []

comfy_api_nodes/nodes_sora.py

@@ -33,7 +33,7 @@ class OpenAIVideoSora2(IO.ComfyNode):
     def define_schema(cls):
         return IO.Schema(
             node_id="OpenAIVideoSora2",
-            display_name="OpenAI Sora - Video (DEPRECATED)",
+            display_name="OpenAI Sora - Video (Deprecated)",
             category="api node/video/Sora",
             description=(
                 "OpenAI video and audio generation.\n\n"

comfy_extras/frame_interpolation_models/film_net.py

@@ -199,9 +199,6 @@ class FILMNet(nn.Module):
     def get_dtype(self):
         return self.extract.extract_sublevels.convs[0][0].conv.weight.dtype
 
-    def memory_used_forward(self, shape, dtype):
-        return 1700 * shape[1] * shape[2] * dtype.itemsize
-
     def _build_warp_grids(self, H, W, device):
         """Pre-compute warp grids for all pyramid levels."""
         if (H, W) in self._warp_grids:

comfy_extras/frame_interpolation_models/ifnet.py

@@ -74,9 +74,6 @@ class IFNet(nn.Module):
     def get_dtype(self):
         return self.encode.cnn0.weight.dtype
 
-    def memory_used_forward(self, shape, dtype):
-        return 300 * shape[1] * shape[2] * dtype.itemsize
-
     def _build_warp_grids(self, H, W, device):
         if (H, W) in self._warp_grids:
             return

comfy_extras/nodes_frame_interpolation.py

@@ -37,7 +37,7 @@ class FrameInterpolationModelLoader(io.ComfyNode):
         model = cls._detect_and_load(sd)
         dtype = torch.float16 if model_management.should_use_fp16(model_management.get_torch_device()) else torch.float32
         model.eval().to(dtype)
-        patcher = comfy.model_patcher.CoreModelPatcher(
+        patcher = comfy.model_patcher.ModelPatcher(
             model,
             load_device=model_management.get_torch_device(),
             offload_device=model_management.unet_offload_device(),
@@ -78,7 +78,7 @@ class FrameInterpolate(io.ComfyNode):
         return io.Schema(
             node_id="FrameInterpolate",
             display_name="Frame Interpolate",
-            category="video",
+            category="image/video",
             search_aliases=["rife", "film", "frame interpolation", "slow motion", "interpolate frames", "vfi"],
             inputs=[
                 FrameInterpolationModel.Input("interp_model"),
@@ -98,13 +98,16 @@ class FrameInterpolate(io.ComfyNode):
         if num_frames < 2 or multiplier < 2:
             return io.NodeOutput(images)
 
+        model_management.load_model_gpu(interp_model)
         device = interp_model.load_device
         dtype = interp_model.model_dtype()
         inference_model = interp_model.model
-        activation_mem = inference_model.memory_used_forward(images.shape, dtype)
-        model_management.load_models_gpu([interp_model], memory_required=activation_mem)
-        align = getattr(inference_model, "pad_align", 1)
+
+        # Free VRAM for inference activations (model weights + ~20x a single frame's worth)
         H, W = images.shape[1], images.shape[2]
+        activation_mem = H * W * 3 * images.element_size() * 20
+        model_management.free_memory(activation_mem, device)
+        align = getattr(inference_model, "pad_align", 1)
 
         # Prepare a single padded frame on device for determining output dimensions
         def prepare_frame(idx):

comfy_extras/nodes_image_compare.py

@@ -11,7 +11,7 @@ class ImageCompare(IO.ComfyNode):
     def define_schema(cls):
         return IO.Schema(
             node_id="ImageCompare",
-            display_name="Compare Images",
+            display_name="Image Compare",
             description="Compares two images side by side with a slider.",
             category="image",
             essentials_category="Image Tools",

comfy_extras/nodes_images.py

@@ -24,7 +24,7 @@ class ImageCrop(IO.ComfyNode):
         return IO.Schema(
             node_id="ImageCrop",
             search_aliases=["trim"],
-            display_name="Crop Image (DEPRECATED)",
+            display_name="Image Crop (Deprecated)",
             category="image/transform",
             is_deprecated=True,
             essentials_category="Image Tools",
@@ -56,7 +56,7 @@ class ImageCropV2(IO.ComfyNode):
         return IO.Schema(
             node_id="ImageCropV2",
             search_aliases=["trim"],
-            display_name="Crop Image",
+            display_name="Image Crop",
             category="image/transform",
             essentials_category="Image Tools",
             has_intermediate_output=True,
@@ -109,7 +109,6 @@ class RepeatImageBatch(IO.ComfyNode):
         return IO.Schema(
             node_id="RepeatImageBatch",
             search_aliases=["duplicate image", "clone image"],
-            display_name="Repeat Image Batch",
             category="image/batch",
             inputs=[
                 IO.Image.Input("image"),
@@ -132,7 +131,6 @@ class ImageFromBatch(IO.ComfyNode):
         return IO.Schema(
             node_id="ImageFromBatch",
             search_aliases=["select image", "pick from batch", "extract image"],
-            display_name="Get Image from Batch",
             category="image/batch",
             inputs=[
                 IO.Image.Input("image"),
@@ -159,8 +157,7 @@ class ImageAddNoise(IO.ComfyNode):
         return IO.Schema(
             node_id="ImageAddNoise",
             search_aliases=["film grain"],
-            display_name="Add Noise to Image",
-            category="image/postprocessing",
+            category="image",
             inputs=[
                 IO.Image.Input("image"),
                 IO.Int.Input(
@@ -262,7 +259,7 @@ class ImageStitch(IO.ComfyNode):
         return IO.Schema(
             node_id="ImageStitch",
             search_aliases=["combine images", "join images", "concatenate images", "side by side"],
-            display_name="Stitch Images",
+            display_name="Image Stitch",
             description="Stitches image2 to image1 in the specified direction.\n"
             "If image2 is not provided, returns image1 unchanged.\n"
             "Optional spacing can be added between images.",
@@ -437,7 +434,6 @@ class ResizeAndPadImage(IO.ComfyNode):
         return IO.Schema(
             node_id="ResizeAndPadImage",
             search_aliases=["fit to size"],
-            display_name="Resize And Pad Image",
             category="image/transform",
             inputs=[
                 IO.Image.Input("image"),
@@ -489,7 +485,6 @@ class SaveSVGNode(IO.ComfyNode):
         return IO.Schema(
             node_id="SaveSVGNode",
             search_aliases=["export vector", "save vector graphics"],
-            display_name="Save SVG",
             description="Save SVG files on disk.",
             category="image/save",
             inputs=[
@@ -596,7 +591,7 @@ class ImageRotate(IO.ComfyNode):
     def define_schema(cls):
         return IO.Schema(
             node_id="ImageRotate",
-            display_name="Rotate Image",
+            display_name="Image Rotate",
             search_aliases=["turn", "flip orientation"],
             category="image/transform",
             essentials_category="Image Tools",
@@ -629,7 +624,6 @@ class ImageFlip(IO.ComfyNode):
         return IO.Schema(
             node_id="ImageFlip",
             search_aliases=["mirror", "reflect"],
-            display_name="Flip Image",
             category="image/transform",
             inputs=[
                 IO.Image.Input("image"),
@@ -656,7 +650,6 @@ class ImageScaleToMaxDimension(IO.ComfyNode):
     def define_schema(cls):
         return IO.Schema(
             node_id="ImageScaleToMaxDimension",
-            display_name="Scale Image to Max Dimension",
             category="image/upscaling",
             inputs=[
                 IO.Image.Input("image"),

comfy_extras/nodes_mask.py

@@ -80,8 +80,7 @@ class ImageCompositeMasked(IO.ComfyNode):
     def define_schema(cls):
         return IO.Schema(
             node_id="ImageCompositeMasked",
-            search_aliases=["overlay", "layer", "paste image", "images composition"],
-            display_name="Image Composite Masked",
+            search_aliases=["paste image", "overlay", "layer"],
             category="image",
             inputs=[
                 IO.Image.Input("destination"),
@@ -202,7 +201,6 @@ class InvertMask(IO.ComfyNode):
         return IO.Schema(
             node_id="InvertMask",
             search_aliases=["reverse mask", "flip mask"],
-            display_name="Invert Mask",
             category="mask",
             inputs=[
                 IO.Mask.Input("mask"),
@@ -224,7 +222,6 @@ class CropMask(IO.ComfyNode):
         return IO.Schema(
             node_id="CropMask",
             search_aliases=["cut mask", "extract mask region", "mask slice"],
-            display_name="Crop Mask",
             category="mask",
             inputs=[
                 IO.Mask.Input("mask"),
@@ -250,8 +247,7 @@ class MaskComposite(IO.ComfyNode):
     def define_schema(cls):
         return IO.Schema(
             node_id="MaskComposite",
-            search_aliases=["combine masks", "blend masks", "layer masks", "masks composition"],
-            display_name="Combine Masks",
+            search_aliases=["combine masks", "blend masks", "layer masks"],
             category="mask",
             inputs=[
                 IO.Mask.Input("destination"),
@@ -302,7 +298,6 @@ class FeatherMask(IO.ComfyNode):
         return IO.Schema(
             node_id="FeatherMask",
             search_aliases=["soft edge mask", "blur mask edges", "gradient mask edge"],
-            display_name="Feather Mask",
             category="mask",
             inputs=[
                 IO.Mask.Input("mask"),

comfy_extras/nodes_morphology.py

@@ -59,8 +59,7 @@ class ImageRGBToYUV(io.ComfyNode):
         return io.Schema(
             node_id="ImageRGBToYUV",
             search_aliases=["color space conversion"],
-            display_name="Image RGB to YUV",
-            category="image/color",
+            category="image/batch",
             inputs=[
                 io.Image.Input("image"),
             ],
@@ -82,8 +81,7 @@ class ImageYUVToRGB(io.ComfyNode):
         return io.Schema(
             node_id="ImageYUVToRGB",
             search_aliases=["color space conversion"],
-            display_name="Image YUV to RGB",
-            category="image/color",
+            category="image/batch",
             inputs=[
                 io.Image.Input("Y"),
                 io.Image.Input("U"),

comfy_extras/nodes_post_processing.py

@@ -20,8 +20,7 @@ class Blend(io.ComfyNode):
     def define_schema(cls):
         return io.Schema(
             node_id="ImageBlend",
-            search_aliases=["mix images"],
-            display_name="Blend Images",
+            display_name="Image Blend",
             category="image/postprocessing",
             essentials_category="Image Tools",
             inputs=[
@@ -225,7 +224,6 @@ class ImageScaleToTotalPixels(io.ComfyNode):
     def define_schema(cls):
         return io.Schema(
             node_id="ImageScaleToTotalPixels",
-            display_name="Scale Image to Total Pixels",
             category="image/upscaling",
             inputs=[
                 io.Image.Input("image"),
@@ -570,7 +568,7 @@ class BatchImagesNode(io.ComfyNode):
         return io.Schema(
             node_id="BatchImagesNode",
             display_name="Batch Images",
-            category="image/batch",
+            category="image",
             essentials_category="Image Tools",
             search_aliases=["batch", "image batch", "batch images", "combine images", "merge images", "stack images"],
             inputs=[
@@ -668,13 +666,12 @@ class ColorTransfer(io.ComfyNode):
     def define_schema(cls):
         return io.Schema(
             node_id="ColorTransfer",
-            display_name="Color Transfer",
             category="image/postprocessing",
             description="Match the colors of one image to another using various algorithms.",
             search_aliases=["color match", "color grading", "color correction", "match colors", "color transform", "mkl", "reinhard", "histogram"],
             inputs=[
                 io.Image.Input("image_target", tooltip="Image(s) to apply the color transform to."),
-                io.Image.Input("image_ref", tooltip="Reference image(s) to match colors to."),
+                io.Image.Input("image_ref", optional=True, tooltip="Reference image(s) to match colors to. If not provided, processing is skipped"),
                 io.Combo.Input("method", options=['reinhard_lab', 'mkl_lab', 'histogram'],),
                 io.DynamicCombo.Input("source_stats",
                     tooltip="per_frame: each frame matched to image_ref individually. uniform: pool stats across all source frames as baseline, match to image_ref. target_frame: use one chosen frame as the baseline for the transform to image_ref, applied uniformly to all frames (preserves relative differences)",

comfy_extras/nodes_primitive.py

@@ -49,7 +49,7 @@ class Int(io.ComfyNode):
             display_name="Int",
             category="utils/primitive",
             inputs=[
-                io.Int.Input("value", min=-sys.maxsize, max=sys.maxsize, control_after_generate=io.ControlAfterGenerate.fixed),
+                io.Int.Input("value", min=-sys.maxsize, max=sys.maxsize, control_after_generate=True),
             ],
             outputs=[io.Int.Output()],
         )

comfy_extras/nodes_video.py

@@ -17,8 +17,7 @@ class SaveWEBM(io.ComfyNode):
         return io.Schema(
             node_id="SaveWEBM",
             search_aliases=["export webm"],
-            display_name="Save WEBM",
-            category="video",
+            category="image/video",
             is_experimental=True,
             inputs=[
                 io.Image.Input("images"),
@@ -73,7 +72,7 @@ class SaveVideo(io.ComfyNode):
             node_id="SaveVideo",
             search_aliases=["export video"],
             display_name="Save Video",
-            category="video",
+            category="image/video",
             essentials_category="Basics",
             description="Saves the input images to your ComfyUI output directory.",
             inputs=[
@@ -122,7 +121,7 @@ class CreateVideo(io.ComfyNode):
             node_id="CreateVideo",
             search_aliases=["images to video"],
             display_name="Create Video",
-            category="video",
+            category="image/video",
             description="Create a video from images.",
             inputs=[
                 io.Image.Input("images", tooltip="The images to create a video from."),
@@ -147,7 +146,7 @@ class GetVideoComponents(io.ComfyNode):
             node_id="GetVideoComponents",
             search_aliases=["extract frames", "split video", "video to images", "demux"],
             display_name="Get Video Components",
-            category="video",
+            category="image/video",
             description="Extracts all components from a video: frames, audio, and framerate.",
             inputs=[
                 io.Video.Input("video", tooltip="The video to extract components from."),
@@ -175,7 +174,7 @@ class LoadVideo(io.ComfyNode):
             node_id="LoadVideo",
             search_aliases=["import video", "open video", "video file"],
             display_name="Load Video",
-            category="video",
+            category="image/video",
             essentials_category="Basics",
             inputs=[
                 io.Combo.Input("file", options=sorted(files), upload=io.UploadType.video),
@@ -217,7 +216,7 @@ class VideoSlice(io.ComfyNode):
                 "frame load cap",
                 "start time",
             ],
-            category="video",
+            category="image/video",
             essentials_category="Video Tools",
             inputs=[
                 io.Video.Input("video"),

extra_model_paths.yaml.example

@@ -28,7 +28,7 @@
 #config for a1111 ui
 #all you have to do is uncomment this (remove the #) and change the base_path to where yours is installed
 
-#a1111:
+#a111:
 #     base_path: path/to/stable-diffusion-webui/
 #     checkpoints: models/Stable-diffusion
 #     configs: models/Stable-diffusion

nodes.py

@@ -1754,49 +1754,57 @@ class LoadImage:
 
         return True
 
-
-class LoadImageMask(LoadImage):
+class LoadImageMask:
     ESSENTIALS_CATEGORY = "Image Tools"
     SEARCH_ALIASES = ["import mask", "alpha mask", "channel mask"]
 
     _color_channels = ["alpha", "red", "green", "blue"]
-
     @classmethod
     def INPUT_TYPES(s):
-        types = super().INPUT_TYPES()
-        return {
-            "required": {
-                **types["required"],
-                "channel": (s._color_channels, )
-            }
-        }
+        input_dir = folder_paths.get_input_directory()
+        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f))]
+        return {"required":
+                    {"image": (sorted(files), {"image_upload": True}),
+                     "channel": (s._color_channels, ), }
+                }
 
     CATEGORY = "mask"
+
     RETURN_TYPES = ("MASK",)
-    FUNCTION = "load_image_mask"
-
-    def load_image_mask(self, image, channel):
-        image_tensor, mask_tensor = super().load_image(image)
+    FUNCTION = "load_image"
+    def load_image(self, image, channel):
+        image_path = folder_paths.get_annotated_filepath(image)
+        i = node_helpers.pillow(Image.open, image_path)
+        i = node_helpers.pillow(ImageOps.exif_transpose, i)
+        if i.getbands() != ("R", "G", "B", "A"):
+            if i.mode == 'I':
+                i = i.point(lambda i: i * (1 / 255))
+            i = i.convert("RGBA")
+        mask = None
         c = channel[0].upper()
-
-        if c == 'A':
-            return (mask_tensor,)
-
-        channel_idx = {'R': 0, 'G': 1, 'B': 2}.get(c, 0)
-
-        if channel_idx < image_tensor.shape[-1]:
-            return (image_tensor[..., channel_idx].clone(),)
+        if c in i.getbands():
+            mask = np.array(i.getchannel(c)).astype(np.float32) / 255.0
+            mask = torch.from_numpy(mask)
+            if c == 'A':
+                mask = 1. - mask
         else:
-            empty_mask = torch.zeros(
-                image_tensor.shape[:-1],
-                dtype=image_tensor.dtype,
-                device=image_tensor.device
-            )
-            return (empty_mask,)
+            mask = torch.zeros((64,64), dtype=torch.float32, device="cpu")
+        return (mask.unsqueeze(0),)
 
     @classmethod
     def IS_CHANGED(s, image, channel):
-        return super().IS_CHANGED(image)
+        image_path = folder_paths.get_annotated_filepath(image)
+        m = hashlib.sha256()
+        with open(image_path, 'rb') as f:
+            m.update(f.read())
+        return m.digest().hex()
+
+    @classmethod
+    def VALIDATE_INPUTS(s, image):
+        if not folder_paths.exists_annotated_filepath(image):
+            return "Invalid image file: {}".format(image)
+
+        return True
 
 
 class LoadImageOutput(LoadImage):
@@ -1887,7 +1895,7 @@ class ImageInvert:
     RETURN_TYPES = ("IMAGE",)
     FUNCTION = "invert"
 
-    CATEGORY = "image/color"
+    CATEGORY = "image"
 
     def invert(self, image):
         s = 1.0 - image
@@ -1903,7 +1911,7 @@ class ImageBatch:
     RETURN_TYPES = ("IMAGE",)
     FUNCTION = "batch"
 
-    CATEGORY = "image/batch"
+    CATEGORY = "image"
     DEPRECATED = True
 
     def batch(self, image1, image2):
@@ -1960,7 +1968,7 @@ class ImagePadForOutpaint:
     RETURN_TYPES = ("IMAGE", "MASK")
     FUNCTION = "expand_image"
 
-    CATEGORY = "image/transform"
+    CATEGORY = "image"
 
     def expand_image(self, image, left, top, right, bottom, feathering):
         d1, d2, d3, d4 = image.size()
@@ -2103,7 +2111,7 @@ NODE_DISPLAY_NAME_MAPPINGS = {
     "ConditioningSetArea": "Conditioning (Set Area)",
     "ConditioningSetAreaPercentage": "Conditioning (Set Area with Percentage)",
     "ConditioningSetMask": "Conditioning (Set Mask)",
-    "ControlNetApply": "Apply ControlNet (DEPRECATED)",
+    "ControlNetApply": "Apply ControlNet (OLD)",
     "ControlNetApplyAdvanced": "Apply ControlNet",
     # Latent
     "VAEEncodeForInpaint": "VAE Encode (for Inpainting)",
@@ -2121,7 +2129,6 @@ NODE_DISPLAY_NAME_MAPPINGS = {
     "LatentFromBatch" : "Latent From Batch",
     "RepeatLatentBatch": "Repeat Latent Batch",
     # Image
-    "EmptyImage": "Empty Image",
     "SaveImage": "Save Image",
     "PreviewImage": "Preview Image",
     "LoadImage": "Load Image",
@@ -2129,15 +2136,15 @@ NODE_DISPLAY_NAME_MAPPINGS = {
     "LoadImageOutput": "Load Image (from Outputs)",
     "ImageScale": "Upscale Image",
     "ImageScaleBy": "Upscale Image By",
-    "ImageInvert": "Invert Image Colors",
+    "ImageInvert": "Invert Image",
     "ImagePadForOutpaint": "Pad Image for Outpainting",
-    "ImageBatch": "Batch Images (DEPRECATED)",
-    "ImageCrop": "Crop Image",
-    "ImageStitch": "Stitch Images",
-    "ImageBlend": "Blend Images",
-    "ImageBlur": "Blur Image",
-    "ImageQuantize": "Quantize Image",
-    "ImageSharpen": "Sharpen Image",
+    "ImageBatch": "Batch Images",
+    "ImageCrop": "Image Crop",
+    "ImageStitch": "Image Stitch",
+    "ImageBlend": "Image Blend",
+    "ImageBlur": "Image Blur",
+    "ImageQuantize": "Image Quantize",
+    "ImageSharpen": "Image Sharpen",
     "ImageScaleToTotalPixels": "Scale Image to Total Pixels",
     "GetImageSize": "Get Image Size",
     # _for_testing