diff --git a/README.md b/README.md index b0f62695b..6d09758c0 100644 --- a/README.md +++ b/README.md @@ -212,6 +212,8 @@ Python 3.14 works but you may encounter issues with the torch compile node. The Python 3.13 is very well supported. If you have trouble with some custom node dependencies on 3.13 you can try 3.12 +torch 2.4 and above is supported but some features might only work on newer versions. We generally recommend using the latest major version of pytorch unless it is less than 2 weeks old. + ### Instructions: Git clone this repo. diff --git a/comfy/context_windows.py b/comfy/context_windows.py index 1e0f86026..2f82d51da 100644 --- a/comfy/context_windows.py +++ b/comfy/context_windows.py @@ -188,6 +188,12 @@ class IndexListContextHandler(ContextHandlerABC): audio_cond = cond_value.cond if audio_cond.ndim > 1 and audio_cond.size(1) == x_in.size(self.dim): new_cond_item[cond_key] = cond_value._copy_with(window.get_tensor(audio_cond, device, dim=1)) + # Handle vace_context (temporal dim is 3) + elif cond_key == "vace_context" and hasattr(cond_value, "cond") and isinstance(cond_value.cond, torch.Tensor): + vace_cond = cond_value.cond + if vace_cond.ndim >= 4 and vace_cond.size(3) == x_in.size(self.dim): + sliced_vace = window.get_tensor(vace_cond, device, dim=3, retain_index_list=self.cond_retain_index_list) + new_cond_item[cond_key] = cond_value._copy_with(sliced_vace) # if has cond that is a Tensor, check if needs to be subset elif hasattr(cond_value, "cond") and isinstance(cond_value.cond, torch.Tensor): if (self.dim < cond_value.cond.ndim and cond_value.cond.size(self.dim) == x_in.size(self.dim)) or \ diff --git a/comfy/k_diffusion/sampling.py b/comfy/k_diffusion/sampling.py index 1ba9edad7..0949dee44 100644 --- a/comfy/k_diffusion/sampling.py +++ b/comfy/k_diffusion/sampling.py @@ -74,6 +74,9 @@ def get_ancestral_step(sigma_from, sigma_to, eta=1.): def default_noise_sampler(x, seed=None): if seed is not None: + if x.device == torch.device("cpu"): + seed += 1 + generator = torch.Generator(device=x.device) generator.manual_seed(seed) else: diff --git a/comfy/ldm/lumina/model.py b/comfy/ldm/lumina/model.py index e80b1c138..afbab2ac7 100644 --- a/comfy/ldm/lumina/model.py +++ b/comfy/ldm/lumina/model.py @@ -491,7 +491,8 @@ class NextDiT(nn.Module): for layer_id in range(n_layers) ] ) - self.norm_final = operation_settings.get("operations").RMSNorm(dim, eps=norm_eps, elementwise_affine=True, device=operation_settings.get("device"), dtype=operation_settings.get("dtype")) + # This norm final is in the lumina 2.0 code but isn't actually used for anything. + # self.norm_final = operation_settings.get("operations").RMSNorm(dim, eps=norm_eps, elementwise_affine=True, device=operation_settings.get("device"), dtype=operation_settings.get("dtype")) self.final_layer = FinalLayer(dim, patch_size, self.out_channels, z_image_modulation=z_image_modulation, operation_settings=operation_settings) if self.pad_tokens_multiple is not None: diff --git a/comfy/model_management.py b/comfy/model_management.py index 1889ab0ac..87baedd73 100644 --- a/comfy/model_management.py +++ b/comfy/model_management.py @@ -1019,8 +1019,8 @@ NUM_STREAMS = 0 if args.async_offload is not None: NUM_STREAMS = args.async_offload else: - # Enable by default on Nvidia - if is_nvidia(): + # Enable by default on Nvidia and AMD + if is_nvidia() or is_amd(): NUM_STREAMS = 2 if args.disable_async_offload: @@ -1126,6 +1126,16 @@ if not args.disable_pinned_memory: PINNING_ALLOWED_TYPES = set(["Parameter", "QuantizedTensor"]) +def discard_cuda_async_error(): + try: + a = torch.tensor([1], dtype=torch.uint8, device=get_torch_device()) + b = torch.tensor([1], dtype=torch.uint8, device=get_torch_device()) + _ = a + b + torch.cuda.synchronize() + except torch.AcceleratorError: + #Dump it! We already know about it from the synchronous return + pass + def pin_memory(tensor): global TOTAL_PINNED_MEMORY if MAX_PINNED_MEMORY <= 0: @@ -1158,6 +1168,9 @@ def pin_memory(tensor): PINNED_MEMORY[ptr] = size TOTAL_PINNED_MEMORY += size return True + else: + logging.warning("Pin error.") + discard_cuda_async_error() return False @@ -1186,6 +1199,9 @@ def unpin_memory(tensor): if len(PINNED_MEMORY) == 0: TOTAL_PINNED_MEMORY = 0 return True + else: + logging.warning("Unpin error.") + discard_cuda_async_error() return False diff --git a/comfy_api/latest/_io.py b/comfy_api/latest/_io.py index 4b14e5ded..ba0b95498 100644 --- a/comfy_api/latest/_io.py +++ b/comfy_api/latest/_io.py @@ -28,9 +28,8 @@ from comfy_api.internal import (_ComfyNodeInternal, _NodeOutputInternal, classpr prune_dict, shallow_clone_class) from ._resources import Resources, ResourcesLocal from comfy_execution.graph_utils import ExecutionBlocker -from ._util import MESH, VOXEL +from ._util import MESH, VOXEL, SVG as _SVG -# from comfy_extras.nodes_images import SVG as SVG_ # NOTE: needs to be moved before can be imported due to circular reference class FolderType(str, Enum): input = "input" @@ -656,7 +655,7 @@ class Video(ComfyTypeIO): @comfytype(io_type="SVG") class SVG(ComfyTypeIO): - Type = Any # TODO: SVG class is defined in comfy_extras/nodes_images.py, causing circular reference; should be moved to somewhere else before referenced directly in v3 + Type = _SVG @comfytype(io_type="LORA_MODEL") class LoraModel(ComfyTypeIO): diff --git a/comfy_api/latest/_util/__init__.py b/comfy_api/latest/_util/__init__.py index fc5431dda..6313eb01b 100644 --- a/comfy_api/latest/_util/__init__.py +++ b/comfy_api/latest/_util/__init__.py @@ -1,5 +1,6 @@ from .video_types import VideoContainer, VideoCodec, VideoComponents from .geometry_types import VOXEL, MESH +from .image_types import SVG __all__ = [ # Utility Types @@ -8,4 +9,5 @@ __all__ = [ "VideoComponents", "VOXEL", "MESH", + "SVG", ] diff --git a/comfy_api/latest/_util/image_types.py b/comfy_api/latest/_util/image_types.py new file mode 100644 index 000000000..f031ed426 --- /dev/null +++ b/comfy_api/latest/_util/image_types.py @@ -0,0 +1,18 @@ +from io import BytesIO + + +class SVG: + """Stores SVG representations via a list of BytesIO objects.""" + + def __init__(self, data: list[BytesIO]): + self.data = data + + def combine(self, other: 'SVG') -> 'SVG': + return SVG(self.data + other.data) + + @staticmethod + def combine_all(svgs: list['SVG']) -> 'SVG': + all_svgs_list: list[BytesIO] = [] + for svg_item in svgs: + all_svgs_list.extend(svg_item.data) + return SVG(all_svgs_list) diff --git a/comfy_api_nodes/apis/gemini_api.py b/comfy_api_nodes/apis/gemini_api.py index f8edc38c9..d81337dae 100644 --- a/comfy_api_nodes/apis/gemini_api.py +++ b/comfy_api_nodes/apis/gemini_api.py @@ -133,6 +133,7 @@ class GeminiImageGenerateContentRequest(BaseModel): systemInstruction: GeminiSystemInstructionContent | None = Field(None) tools: list[GeminiTool] | None = Field(None) videoMetadata: GeminiVideoMetadata | None = Field(None) + uploadImagesToStorage: bool = Field(True) class GeminiGenerateContentRequest(BaseModel): diff --git a/comfy_api_nodes/apis/kling_api.py b/comfy_api_nodes/apis/kling_api.py index 80a758466..bf54ede3e 100644 --- a/comfy_api_nodes/apis/kling_api.py +++ b/comfy_api_nodes/apis/kling_api.py @@ -102,3 +102,12 @@ class ImageToVideoWithAudioRequest(BaseModel): prompt: str = Field(...) mode: str = Field("pro") sound: str = Field(..., description="'on' or 'off'") + + +class MotionControlRequest(BaseModel): + prompt: str = Field(...) + image_url: str = Field(...) + video_url: str = Field(...) + keep_original_sound: str = Field(...) + character_orientation: str = Field(...) + mode: str = Field(..., description="'pro' or 'std'") diff --git a/comfy_api_nodes/nodes_bytedance.py b/comfy_api_nodes/nodes_bytedance.py index 636cc1265..d4a2cfae6 100644 --- a/comfy_api_nodes/nodes_bytedance.py +++ b/comfy_api_nodes/nodes_bytedance.py @@ -229,6 +229,7 @@ class ByteDanceImageEditNode(IO.ComfyNode): IO.Hidden.unique_id, ], is_api_node=True, + is_deprecated=True, ) @classmethod @@ -269,7 +270,7 @@ class ByteDanceSeedreamNode(IO.ComfyNode): def define_schema(cls): return IO.Schema( node_id="ByteDanceSeedreamNode", - display_name="ByteDance Seedream 4", + display_name="ByteDance Seedream 4.5", category="api node/image/ByteDance", description="Unified text-to-image generation and precise single-sentence editing at up to 4K resolution.", inputs=[ diff --git a/comfy_api_nodes/nodes_gemini.py b/comfy_api_nodes/nodes_gemini.py index ad0f4b4d1..e8ed7e797 100644 --- a/comfy_api_nodes/nodes_gemini.py +++ b/comfy_api_nodes/nodes_gemini.py @@ -34,6 +34,7 @@ from comfy_api_nodes.util import ( ApiEndpoint, audio_to_base64_string, bytesio_to_image_tensor, + download_url_to_image_tensor, get_number_of_images, sync_op, tensor_to_base64_string, @@ -141,9 +142,11 @@ def get_parts_by_type(response: GeminiGenerateContentResponse, part_type: Litera ) parts = [] for part in response.candidates[0].content.parts: - if part_type == "text" and hasattr(part, "text") and part.text: + if part_type == "text" and part.text: parts.append(part) - elif hasattr(part, "inlineData") and part.inlineData and part.inlineData.mimeType == part_type: + elif part.inlineData and part.inlineData.mimeType == part_type: + parts.append(part) + elif part.fileData and part.fileData.mimeType == part_type: parts.append(part) # Skip parts that don't match the requested type return parts @@ -163,12 +166,15 @@ def get_text_from_response(response: GeminiGenerateContentResponse) -> str: return "\n".join([part.text for part in parts]) -def get_image_from_response(response: GeminiGenerateContentResponse) -> Input.Image: +async def get_image_from_response(response: GeminiGenerateContentResponse) -> Input.Image: image_tensors: list[Input.Image] = [] parts = get_parts_by_type(response, "image/png") for part in parts: - image_data = base64.b64decode(part.inlineData.data) - returned_image = bytesio_to_image_tensor(BytesIO(image_data)) + if part.inlineData: + image_data = base64.b64decode(part.inlineData.data) + returned_image = bytesio_to_image_tensor(BytesIO(image_data)) + else: + returned_image = await download_url_to_image_tensor(part.fileData.fileUri) image_tensors.append(returned_image) if len(image_tensors) == 0: return torch.zeros((1, 1024, 1024, 4)) @@ -596,7 +602,7 @@ class GeminiImage(IO.ComfyNode): response = await sync_op( cls, - endpoint=ApiEndpoint(path=f"{GEMINI_BASE_ENDPOINT}/{model}", method="POST"), + ApiEndpoint(path=f"/proxy/vertexai/gemini/{model}", method="POST"), data=GeminiImageGenerateContentRequest( contents=[ GeminiContent(role=GeminiRole.user, parts=parts), @@ -610,7 +616,7 @@ class GeminiImage(IO.ComfyNode): response_model=GeminiGenerateContentResponse, price_extractor=calculate_tokens_price, ) - return IO.NodeOutput(get_image_from_response(response), get_text_from_response(response)) + return IO.NodeOutput(await get_image_from_response(response), get_text_from_response(response)) class GeminiImage2(IO.ComfyNode): @@ -729,7 +735,7 @@ class GeminiImage2(IO.ComfyNode): response = await sync_op( cls, - ApiEndpoint(path=f"{GEMINI_BASE_ENDPOINT}/{model}", method="POST"), + ApiEndpoint(path=f"/proxy/vertexai/gemini/{model}", method="POST"), data=GeminiImageGenerateContentRequest( contents=[ GeminiContent(role=GeminiRole.user, parts=parts), @@ -743,7 +749,7 @@ class GeminiImage2(IO.ComfyNode): response_model=GeminiGenerateContentResponse, price_extractor=calculate_tokens_price, ) - return IO.NodeOutput(get_image_from_response(response), get_text_from_response(response)) + return IO.NodeOutput(await get_image_from_response(response), get_text_from_response(response)) class GeminiExtension(ComfyExtension): diff --git a/comfy_api_nodes/nodes_kling.py b/comfy_api_nodes/nodes_kling.py index 1a6364fa0..58259e029 100644 --- a/comfy_api_nodes/nodes_kling.py +++ b/comfy_api_nodes/nodes_kling.py @@ -51,6 +51,7 @@ from comfy_api_nodes.apis import ( ) from comfy_api_nodes.apis.kling_api import ( ImageToVideoWithAudioRequest, + MotionControlRequest, OmniImageParamImage, OmniParamImage, OmniParamVideo, @@ -858,7 +859,7 @@ class OmniProFirstLastFrameNode(IO.ComfyNode): tooltip="A text prompt describing the video content. " "This can include both positive and negative descriptions.", ), - IO.Combo.Input("duration", options=["5", "10"]), + IO.Int.Input("duration", default=5, min=3, max=10, display_mode=IO.NumberDisplay.slider), IO.Image.Input("first_frame"), IO.Image.Input( "end_frame", @@ -897,6 +898,10 @@ class OmniProFirstLastFrameNode(IO.ComfyNode): validate_string(prompt, min_length=1, max_length=2500) if end_frame is not None and reference_images is not None: raise ValueError("The 'end_frame' input cannot be used simultaneously with 'reference_images'.") + if duration not in (5, 10) and end_frame is None and reference_images is None: + raise ValueError( + "Duration is only supported for 5 or 10 seconds if there is no end frame or reference images." + ) validate_image_dimensions(first_frame, min_width=300, min_height=300) validate_image_aspect_ratio(first_frame, (1, 2.5), (2.5, 1)) image_list: list[OmniParamImage] = [ @@ -2159,6 +2164,91 @@ class ImageToVideoWithAudio(IO.ComfyNode): return IO.NodeOutput(await download_url_to_video_output(final_response.data.task_result.videos[0].url)) +class MotionControl(IO.ComfyNode): + + @classmethod + def define_schema(cls) -> IO.Schema: + return IO.Schema( + node_id="KlingMotionControl", + display_name="Kling Motion Control", + category="api node/video/Kling", + inputs=[ + IO.String.Input("prompt", multiline=True), + IO.Image.Input("reference_image"), + IO.Video.Input( + "reference_video", + tooltip="Motion reference video used to drive movement/expression.\n" + "Duration limits depend on character_orientation:\n" + " - image: 3–10s (max 10s)\n" + " - video: 3–30s (max 30s)", + ), + IO.Boolean.Input("keep_original_sound", default=True), + IO.Combo.Input( + "character_orientation", + options=["video", "image"], + tooltip="Controls where the character's facing/orientation comes from.\n" + "video: movements, expressions, camera moves, and orientation " + "follow the motion reference video (other details via prompt).\n" + "image: movements and expressions still follow the motion reference video, " + "but the character orientation matches the reference image (camera/other details via prompt).", + ), + IO.Combo.Input("mode", options=["pro", "std"]), + ], + outputs=[ + IO.Video.Output(), + ], + hidden=[ + IO.Hidden.auth_token_comfy_org, + IO.Hidden.api_key_comfy_org, + IO.Hidden.unique_id, + ], + is_api_node=True, + ) + + @classmethod + async def execute( + cls, + prompt: str, + reference_image: Input.Image, + reference_video: Input.Video, + keep_original_sound: bool, + character_orientation: str, + mode: str, + ) -> IO.NodeOutput: + validate_string(prompt, max_length=2500) + validate_image_dimensions(reference_image, min_width=340, min_height=340) + validate_image_aspect_ratio(reference_image, (1, 2.5), (2.5, 1)) + if character_orientation == "image": + validate_video_duration(reference_video, min_duration=3, max_duration=10) + else: + validate_video_duration(reference_video, min_duration=3, max_duration=30) + validate_video_dimensions(reference_video, min_width=340, min_height=340, max_width=3850, max_height=3850) + response = await sync_op( + cls, + ApiEndpoint(path="/proxy/kling/v1/videos/motion-control", method="POST"), + response_model=TaskStatusResponse, + data=MotionControlRequest( + prompt=prompt, + image_url=(await upload_images_to_comfyapi(cls, reference_image))[0], + video_url=await upload_video_to_comfyapi(cls, reference_video), + keep_original_sound="yes" if keep_original_sound else "no", + character_orientation=character_orientation, + mode=mode, + ), + ) + if response.code: + raise RuntimeError( + f"Kling request failed. Code: {response.code}, Message: {response.message}, Data: {response.data}" + ) + final_response = await poll_op( + cls, + ApiEndpoint(path=f"/proxy/kling/v1/videos/motion-control/{response.data.task_id}"), + response_model=TaskStatusResponse, + status_extractor=lambda r: (r.data.task_status if r.data else None), + ) + return IO.NodeOutput(await download_url_to_video_output(final_response.data.task_result.videos[0].url)) + + class KlingExtension(ComfyExtension): @override async def get_node_list(self) -> list[type[IO.ComfyNode]]: @@ -2184,6 +2274,7 @@ class KlingExtension(ComfyExtension): OmniProImageNode, TextToVideoWithAudio, ImageToVideoWithAudio, + MotionControl, ] diff --git a/comfy_api_nodes/nodes_veo2.py b/comfy_api_nodes/nodes_veo2.py index e165b8380..13a6bfd91 100644 --- a/comfy_api_nodes/nodes_veo2.py +++ b/comfy_api_nodes/nodes_veo2.py @@ -168,6 +168,8 @@ class VeoVideoGenerationNode(IO.ComfyNode): # Only add generateAudio for Veo 3 models if model.find("veo-2.0") == -1: parameters["generateAudio"] = generate_audio + # force "enhance_prompt" to True for Veo3 models + parameters["enhancePrompt"] = True initial_response = await sync_op( cls, @@ -291,7 +293,7 @@ class Veo3VideoGenerationNode(VeoVideoGenerationNode): IO.Boolean.Input( "enhance_prompt", default=True, - tooltip="Whether to enhance the prompt with AI assistance", + tooltip="This parameter is deprecated and ignored.", optional=True, ), IO.Combo.Input( diff --git a/comfy_api_nodes/util/client.py b/comfy_api_nodes/util/client.py index bf37cba5f..f372ec7b5 100644 --- a/comfy_api_nodes/util/client.py +++ b/comfy_api_nodes/util/client.py @@ -430,9 +430,9 @@ def _display_text( if status: display_lines.append(f"Status: {status.capitalize() if isinstance(status, str) else status}") if price is not None: - p = f"{float(price):,.4f}".rstrip("0").rstrip(".") + p = f"{float(price) * 211:,.1f}".rstrip("0").rstrip(".") if p != "0": - display_lines.append(f"Price: ${p}") + display_lines.append(f"Price: {p} credits") if text is not None: display_lines.append(text) if display_lines: diff --git a/comfy_extras/nodes_custom_sampler.py b/comfy_extras/nodes_custom_sampler.py index 7ee4caac1..f19adf4b9 100644 --- a/comfy_extras/nodes_custom_sampler.py +++ b/comfy_extras/nodes_custom_sampler.py @@ -9,6 +9,7 @@ import comfy.utils import node_helpers from typing_extensions import override from comfy_api.latest import ComfyExtension, io +import re class BasicScheduler(io.ComfyNode): @@ -760,8 +761,12 @@ class SamplerCustom(io.ComfyNode): out = latent.copy() out["samples"] = samples if "x0" in x0_output: + x0_out = model.model.process_latent_out(x0_output["x0"].cpu()) + if samples.is_nested: + latent_shapes = [x.shape for x in samples.unbind()] + x0_out = comfy.nested_tensor.NestedTensor(comfy.utils.unpack_latents(x0_out, latent_shapes)) out_denoised = latent.copy() - out_denoised["samples"] = model.model.process_latent_out(x0_output["x0"].cpu()) + out_denoised["samples"] = x0_out else: out_denoised = out return io.NodeOutput(out, out_denoised) @@ -948,8 +953,12 @@ class SamplerCustomAdvanced(io.ComfyNode): out = latent.copy() out["samples"] = samples if "x0" in x0_output: + x0_out = guider.model_patcher.model.process_latent_out(x0_output["x0"].cpu()) + if samples.is_nested: + latent_shapes = [x.shape for x in samples.unbind()] + x0_out = comfy.nested_tensor.NestedTensor(comfy.utils.unpack_latents(x0_out, latent_shapes)) out_denoised = latent.copy() - out_denoised["samples"] = guider.model_patcher.model.process_latent_out(x0_output["x0"].cpu()) + out_denoised["samples"] = x0_out else: out_denoised = out return io.NodeOutput(out, out_denoised) @@ -1005,6 +1014,25 @@ class AddNoise(io.ComfyNode): add_noise = execute +class ManualSigmas(io.ComfyNode): + @classmethod + def define_schema(cls): + return io.Schema( + node_id="ManualSigmas", + category="_for_testing/custom_sampling", + is_experimental=True, + inputs=[ + io.String.Input("sigmas", default="1, 0.5", multiline=False) + ], + outputs=[io.Sigmas.Output()] + ) + + @classmethod + def execute(cls, sigmas) -> io.NodeOutput: + sigmas = re.findall(r"[-+]?(?:\d*\.*\d+)", sigmas) + sigmas = [float(i) for i in sigmas] + sigmas = torch.FloatTensor(sigmas) + return io.NodeOutput(sigmas) class CustomSamplersExtension(ComfyExtension): @override @@ -1044,6 +1072,7 @@ class CustomSamplersExtension(ComfyExtension): DisableNoise, AddNoise, SamplerCustomAdvanced, + ManualSigmas, ] diff --git a/comfy_extras/nodes_dataset.py b/comfy_extras/nodes_dataset.py index 513aecf3a..5ef851bd0 100644 --- a/comfy_extras/nodes_dataset.py +++ b/comfy_extras/nodes_dataset.py @@ -667,16 +667,19 @@ class ResizeImagesByLongerEdgeNode(ImageProcessingNode): @classmethod def _process(cls, image, longer_edge): - img = tensor_to_pil(image) - w, h = img.size - if w > h: - new_w = longer_edge - new_h = int(h * (longer_edge / w)) - else: - new_h = longer_edge - new_w = int(w * (longer_edge / h)) - img = img.resize((new_w, new_h), Image.Resampling.LANCZOS) - return pil_to_tensor(img) + resized_images = [] + for image_i in image: + img = tensor_to_pil(image_i) + w, h = img.size + if w > h: + new_w = longer_edge + new_h = int(h * (longer_edge / w)) + else: + new_h = longer_edge + new_w = int(w * (longer_edge / h)) + img = img.resize((new_w, new_h), Image.Resampling.LANCZOS) + resized_images.append(pil_to_tensor(img)) + return torch.cat(resized_images, dim=0) class CenterCropImagesNode(ImageProcessingNode): diff --git a/comfy_extras/nodes_images.py b/comfy_extras/nodes_images.py index 392aea32c..ce21caade 100644 --- a/comfy_extras/nodes_images.py +++ b/comfy_extras/nodes_images.py @@ -2,280 +2,231 @@ from __future__ import annotations import nodes import folder_paths -from comfy.cli_args import args -from PIL import Image -from PIL.PngImagePlugin import PngInfo - -import numpy as np import json import os import re -from io import BytesIO -from inspect import cleandoc import torch import comfy.utils -from comfy.comfy_types import FileLocator, IO from server import PromptServer +from comfy_api.latest import ComfyExtension, IO, UI +from typing_extensions import override + +SVG = IO.SVG.Type # TODO: temporary solution for backward compatibility, will be removed later. MAX_RESOLUTION = nodes.MAX_RESOLUTION -class ImageCrop: +class ImageCrop(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return {"required": { "image": ("IMAGE",), - "width": ("INT", {"default": 512, "min": 1, "max": MAX_RESOLUTION, "step": 1}), - "height": ("INT", {"default": 512, "min": 1, "max": MAX_RESOLUTION, "step": 1}), - "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 1}), - "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 1}), - }} - RETURN_TYPES = ("IMAGE",) - FUNCTION = "crop" + def define_schema(cls): + return IO.Schema( + node_id="ImageCrop", + display_name="Image Crop", + category="image/transform", + inputs=[ + IO.Image.Input("image"), + IO.Int.Input("width", default=512, min=1, max=nodes.MAX_RESOLUTION, step=1), + IO.Int.Input("height", default=512, min=1, max=nodes.MAX_RESOLUTION, step=1), + IO.Int.Input("x", default=0, min=0, max=nodes.MAX_RESOLUTION, step=1), + IO.Int.Input("y", default=0, min=0, max=nodes.MAX_RESOLUTION, step=1), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image/transform" - - def crop(self, image, width, height, x, y): + @classmethod + def execute(cls, image, width, height, x, y) -> IO.NodeOutput: x = min(x, image.shape[2] - 1) y = min(y, image.shape[1] - 1) to_x = width + x to_y = height + y img = image[:,y:to_y, x:to_x, :] - return (img,) + return IO.NodeOutput(img) -class RepeatImageBatch: + crop = execute # TODO: remove + + +class RepeatImageBatch(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return {"required": { "image": ("IMAGE",), - "amount": ("INT", {"default": 1, "min": 1, "max": 4096}), - }} - RETURN_TYPES = ("IMAGE",) - FUNCTION = "repeat" + def define_schema(cls): + return IO.Schema( + node_id="RepeatImageBatch", + category="image/batch", + inputs=[ + IO.Image.Input("image"), + IO.Int.Input("amount", default=1, min=1, max=4096), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image/batch" - - def repeat(self, image, amount): + @classmethod + def execute(cls, image, amount) -> IO.NodeOutput: s = image.repeat((amount, 1,1,1)) - return (s,) + return IO.NodeOutput(s) -class ImageFromBatch: + repeat = execute # TODO: remove + + +class ImageFromBatch(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return {"required": { "image": ("IMAGE",), - "batch_index": ("INT", {"default": 0, "min": 0, "max": 4095}), - "length": ("INT", {"default": 1, "min": 1, "max": 4096}), - }} - RETURN_TYPES = ("IMAGE",) - FUNCTION = "frombatch" + def define_schema(cls): + return IO.Schema( + node_id="ImageFromBatch", + category="image/batch", + inputs=[ + IO.Image.Input("image"), + IO.Int.Input("batch_index", default=0, min=0, max=4095), + IO.Int.Input("length", default=1, min=1, max=4096), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image/batch" - - def frombatch(self, image, batch_index, length): + @classmethod + def execute(cls, image, batch_index, length) -> IO.NodeOutput: s_in = image batch_index = min(s_in.shape[0] - 1, batch_index) length = min(s_in.shape[0] - batch_index, length) s = s_in[batch_index:batch_index + length].clone() - return (s,) + return IO.NodeOutput(s) + + frombatch = execute # TODO: remove -class ImageAddNoise: +class ImageAddNoise(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return {"required": { "image": ("IMAGE",), - "seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff, "control_after_generate": True, "tooltip": "The random seed used for creating the noise."}), - "strength": ("FLOAT", {"default": 0.5, "min": 0.0, "max": 1.0, "step": 0.01}), - }} - RETURN_TYPES = ("IMAGE",) - FUNCTION = "repeat" + def define_schema(cls): + return IO.Schema( + node_id="ImageAddNoise", + category="image", + inputs=[ + IO.Image.Input("image"), + IO.Int.Input( + "seed", + default=0, + min=0, + max=0xFFFFFFFFFFFFFFFF, + control_after_generate=True, + tooltip="The random seed used for creating the noise.", + ), + IO.Float.Input("strength", default=0.5, min=0.0, max=1.0, step=0.01), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image" - - def repeat(self, image, seed, strength): + @classmethod + def execute(cls, image, seed, strength) -> IO.NodeOutput: generator = torch.manual_seed(seed) s = torch.clip((image + strength * torch.randn(image.size(), generator=generator, device="cpu").to(image)), min=0.0, max=1.0) - return (s,) + return IO.NodeOutput(s) -class SaveAnimatedWEBP: - def __init__(self): - self.output_dir = folder_paths.get_output_directory() - self.type = "output" - self.prefix_append = "" + repeat = execute # TODO: remove - methods = {"default": 4, "fastest": 0, "slowest": 6} - @classmethod - def INPUT_TYPES(s): - return {"required": - {"images": ("IMAGE", ), - "filename_prefix": ("STRING", {"default": "ComfyUI"}), - "fps": ("FLOAT", {"default": 6.0, "min": 0.01, "max": 1000.0, "step": 0.01}), - "lossless": ("BOOLEAN", {"default": True}), - "quality": ("INT", {"default": 80, "min": 0, "max": 100}), - "method": (list(s.methods.keys()),), - # "num_frames": ("INT", {"default": 0, "min": 0, "max": 8192}), - }, - "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"}, - } - RETURN_TYPES = () - FUNCTION = "save_images" - - OUTPUT_NODE = True - - CATEGORY = "image/animation" - - def save_images(self, images, fps, filename_prefix, lossless, quality, method, num_frames=0, prompt=None, extra_pnginfo=None): - method = self.methods.get(method) - filename_prefix += self.prefix_append - full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir, images[0].shape[1], images[0].shape[0]) - results: list[FileLocator] = [] - pil_images = [] - for image in images: - i = 255. * image.cpu().numpy() - img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8)) - pil_images.append(img) - - metadata = pil_images[0].getexif() - if not args.disable_metadata: - if prompt is not None: - metadata[0x0110] = "prompt:{}".format(json.dumps(prompt)) - if extra_pnginfo is not None: - inital_exif = 0x010f - for x in extra_pnginfo: - metadata[inital_exif] = "{}:{}".format(x, json.dumps(extra_pnginfo[x])) - inital_exif -= 1 - - if num_frames == 0: - num_frames = len(pil_images) - - c = len(pil_images) - for i in range(0, c, num_frames): - file = f"{filename}_{counter:05}_.webp" - pil_images[i].save(os.path.join(full_output_folder, file), save_all=True, duration=int(1000.0/fps), append_images=pil_images[i + 1:i + num_frames], exif=metadata, lossless=lossless, quality=quality, method=method) - results.append({ - "filename": file, - "subfolder": subfolder, - "type": self.type - }) - counter += 1 - - animated = num_frames != 1 - return { "ui": { "images": results, "animated": (animated,) } } - -class SaveAnimatedPNG: - def __init__(self): - self.output_dir = folder_paths.get_output_directory() - self.type = "output" - self.prefix_append = "" +class SaveAnimatedWEBP(IO.ComfyNode): + COMPRESS_METHODS = {"default": 4, "fastest": 0, "slowest": 6} @classmethod - def INPUT_TYPES(s): - return {"required": - {"images": ("IMAGE", ), - "filename_prefix": ("STRING", {"default": "ComfyUI"}), - "fps": ("FLOAT", {"default": 6.0, "min": 0.01, "max": 1000.0, "step": 0.01}), - "compress_level": ("INT", {"default": 4, "min": 0, "max": 9}) - }, - "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"}, - } + def define_schema(cls): + return IO.Schema( + node_id="SaveAnimatedWEBP", + category="image/animation", + inputs=[ + IO.Image.Input("images"), + IO.String.Input("filename_prefix", default="ComfyUI"), + IO.Float.Input("fps", default=6.0, min=0.01, max=1000.0, step=0.01), + IO.Boolean.Input("lossless", default=True), + IO.Int.Input("quality", default=80, min=0, max=100), + IO.Combo.Input("method", options=list(cls.COMPRESS_METHODS.keys())), + # "num_frames": ("INT", {"default": 0, "min": 0, "max": 8192}), + ], + hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo], + is_output_node=True, + ) - RETURN_TYPES = () - FUNCTION = "save_images" + @classmethod + def execute(cls, images, fps, filename_prefix, lossless, quality, method, num_frames=0) -> IO.NodeOutput: + return IO.NodeOutput( + ui=UI.ImageSaveHelper.get_save_animated_webp_ui( + images=images, + filename_prefix=filename_prefix, + cls=cls, + fps=fps, + lossless=lossless, + quality=quality, + method=cls.COMPRESS_METHODS.get(method) + ) + ) - OUTPUT_NODE = True - - CATEGORY = "image/animation" - - def save_images(self, images, fps, compress_level, filename_prefix="ComfyUI", prompt=None, extra_pnginfo=None): - filename_prefix += self.prefix_append - full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir, images[0].shape[1], images[0].shape[0]) - results = list() - pil_images = [] - for image in images: - i = 255. * image.cpu().numpy() - img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8)) - pil_images.append(img) - - metadata = None - if not args.disable_metadata: - metadata = PngInfo() - if prompt is not None: - metadata.add(b"comf", "prompt".encode("latin-1", "strict") + b"\0" + json.dumps(prompt).encode("latin-1", "strict"), after_idat=True) - if extra_pnginfo is not None: - for x in extra_pnginfo: - metadata.add(b"comf", x.encode("latin-1", "strict") + b"\0" + json.dumps(extra_pnginfo[x]).encode("latin-1", "strict"), after_idat=True) - - file = f"{filename}_{counter:05}_.png" - pil_images[0].save(os.path.join(full_output_folder, file), pnginfo=metadata, compress_level=compress_level, save_all=True, duration=int(1000.0/fps), append_images=pil_images[1:]) - results.append({ - "filename": file, - "subfolder": subfolder, - "type": self.type - }) - - return { "ui": { "images": results, "animated": (True,)} } - -class SVG: - """ - Stores SVG representations via a list of BytesIO objects. - """ - def __init__(self, data: list[BytesIO]): - self.data = data - - def combine(self, other: 'SVG') -> 'SVG': - return SVG(self.data + other.data) - - @staticmethod - def combine_all(svgs: list['SVG']) -> 'SVG': - all_svgs_list: list[BytesIO] = [] - for svg_item in svgs: - all_svgs_list.extend(svg_item.data) - return SVG(all_svgs_list) + save_images = execute # TODO: remove -class ImageStitch: +class SaveAnimatedPNG(IO.ComfyNode): + + @classmethod + def define_schema(cls): + return IO.Schema( + node_id="SaveAnimatedPNG", + category="image/animation", + inputs=[ + IO.Image.Input("images"), + IO.String.Input("filename_prefix", default="ComfyUI"), + IO.Float.Input("fps", default=6.0, min=0.01, max=1000.0, step=0.01), + IO.Int.Input("compress_level", default=4, min=0, max=9), + ], + hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo], + is_output_node=True, + ) + + @classmethod + def execute(cls, images, fps, compress_level, filename_prefix="ComfyUI") -> IO.NodeOutput: + return IO.NodeOutput( + ui=UI.ImageSaveHelper.get_save_animated_png_ui( + images=images, + filename_prefix=filename_prefix, + cls=cls, + fps=fps, + compress_level=compress_level, + ) + ) + + save_images = execute # TODO: remove + + +class ImageStitch(IO.ComfyNode): """Upstreamed from https://github.com/kijai/ComfyUI-KJNodes""" @classmethod - def INPUT_TYPES(s): - return { - "required": { - "image1": ("IMAGE",), - "direction": (["right", "down", "left", "up"], {"default": "right"}), - "match_image_size": ("BOOLEAN", {"default": True}), - "spacing_width": ( - "INT", - {"default": 0, "min": 0, "max": 1024, "step": 2}, - ), - "spacing_color": ( - ["white", "black", "red", "green", "blue"], - {"default": "white"}, - ), - }, - "optional": { - "image2": ("IMAGE",), - }, - } + def define_schema(cls): + return IO.Schema( + node_id="ImageStitch", + 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.", + category="image/transform", + inputs=[ + IO.Image.Input("image1"), + IO.Combo.Input("direction", options=["right", "down", "left", "up"], default="right"), + IO.Boolean.Input("match_image_size", default=True), + IO.Int.Input("spacing_width", default=0, min=0, max=1024, step=2), + IO.Combo.Input("spacing_color", options=["white", "black", "red", "green", "blue"], default="white"), + IO.Image.Input("image2", optional=True), + ], + outputs=[IO.Image.Output()], + ) - RETURN_TYPES = ("IMAGE",) - FUNCTION = "stitch" - CATEGORY = "image/transform" - DESCRIPTION = """ -Stitches image2 to image1 in the specified direction. -If image2 is not provided, returns image1 unchanged. -Optional spacing can be added between images. -""" - - def stitch( - self, + @classmethod + def execute( + cls, image1, direction, match_image_size, spacing_width, spacing_color, image2=None, - ): + ) -> IO.NodeOutput: if image2 is None: - return (image1,) + return IO.NodeOutput(image1) # Handle batch size differences if image1.shape[0] != image2.shape[0]: @@ -412,36 +363,30 @@ Optional spacing can be added between images. images.insert(1, spacing) concat_dim = 2 if direction in ["left", "right"] else 1 - return (torch.cat(images, dim=concat_dim),) + return IO.NodeOutput(torch.cat(images, dim=concat_dim)) + + stitch = execute # TODO: remove + + +class ResizeAndPadImage(IO.ComfyNode): -class ResizeAndPadImage: @classmethod - def INPUT_TYPES(cls): - return { - "required": { - "image": ("IMAGE",), - "target_width": ("INT", { - "default": 512, - "min": 1, - "max": MAX_RESOLUTION, - "step": 1 - }), - "target_height": ("INT", { - "default": 512, - "min": 1, - "max": MAX_RESOLUTION, - "step": 1 - }), - "padding_color": (["white", "black"],), - "interpolation": (["area", "bicubic", "nearest-exact", "bilinear", "lanczos"],), - } - } + def define_schema(cls): + return IO.Schema( + node_id="ResizeAndPadImage", + category="image/transform", + inputs=[ + IO.Image.Input("image"), + IO.Int.Input("target_width", default=512, min=1, max=nodes.MAX_RESOLUTION, step=1), + IO.Int.Input("target_height", default=512, min=1, max=nodes.MAX_RESOLUTION, step=1), + IO.Combo.Input("padding_color", options=["white", "black"]), + IO.Combo.Input("interpolation", options=["area", "bicubic", "nearest-exact", "bilinear", "lanczos"]), + ], + outputs=[IO.Image.Output()], + ) - RETURN_TYPES = ("IMAGE",) - FUNCTION = "resize_and_pad" - CATEGORY = "image/transform" - - def resize_and_pad(self, image, target_width, target_height, padding_color, interpolation): + @classmethod + def execute(cls, image, target_width, target_height, padding_color, interpolation) -> IO.NodeOutput: batch_size, orig_height, orig_width, channels = image.shape scale_w = target_width / orig_width @@ -469,52 +414,47 @@ class ResizeAndPadImage: padded[:, :, y_offset:y_offset + new_height, x_offset:x_offset + new_width] = resized output = padded.permute(0, 2, 3, 1) - return (output,) + return IO.NodeOutput(output) -class SaveSVGNode: - """ - Save SVG files on disk. - """ + resize_and_pad = execute # TODO: remove - def __init__(self): - self.output_dir = folder_paths.get_output_directory() - self.type = "output" - self.prefix_append = "" - RETURN_TYPES = () - DESCRIPTION = cleandoc(__doc__ or "") # Handle potential None value - FUNCTION = "save_svg" - CATEGORY = "image/save" # Changed - OUTPUT_NODE = True +class SaveSVGNode(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return { - "required": { - "svg": ("SVG",), # Changed - "filename_prefix": ("STRING", {"default": "svg/ComfyUI", "tooltip": "The prefix for the file to save. This may include formatting information such as %date:yyyy-MM-dd% or %Empty Latent Image.width% to include values from nodes."}) - }, - "hidden": { - "prompt": "PROMPT", - "extra_pnginfo": "EXTRA_PNGINFO" - } - } + def define_schema(cls): + return IO.Schema( + node_id="SaveSVGNode", + description="Save SVG files on disk.", + category="image/save", + inputs=[ + IO.SVG.Input("svg"), + IO.String.Input( + "filename_prefix", + default="svg/ComfyUI", + tooltip="The prefix for the file to save. This may include formatting information such as %date:yyyy-MM-dd% or %Empty Latent Image.width% to include values from nodes.", + ), + ], + hidden=[IO.Hidden.prompt, IO.Hidden.extra_pnginfo], + is_output_node=True, + ) - def save_svg(self, svg: SVG, filename_prefix="svg/ComfyUI", prompt=None, extra_pnginfo=None): - filename_prefix += self.prefix_append - full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir) - results = list() + @classmethod + def execute(cls, svg: IO.SVG.Type, filename_prefix="svg/ComfyUI") -> IO.NodeOutput: + full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, folder_paths.get_output_directory()) + results: list[UI.SavedResult] = [] # Prepare metadata JSON metadata_dict = {} - if prompt is not None: - metadata_dict["prompt"] = prompt - if extra_pnginfo is not None: - metadata_dict.update(extra_pnginfo) + if cls.hidden.prompt is not None: + metadata_dict["prompt"] = cls.hidden.prompt + if cls.hidden.extra_pnginfo is not None: + metadata_dict.update(cls.hidden.extra_pnginfo) # Convert metadata to JSON string metadata_json = json.dumps(metadata_dict, indent=2) if metadata_dict else None + for batch_number, svg_bytes in enumerate(svg.data): filename_with_batch_num = filename.replace("%batch_num%", str(batch_number)) file = f"{filename_with_batch_num}_{counter:05}_.svg" @@ -544,57 +484,64 @@ class SaveSVGNode: with open(os.path.join(full_output_folder, file), 'wb') as svg_file: svg_file.write(svg_content.encode('utf-8')) - results.append({ - "filename": file, - "subfolder": subfolder, - "type": self.type - }) + results.append(UI.SavedResult(filename=file, subfolder=subfolder, type=IO.FolderType.output)) counter += 1 - return { "ui": { "images": results } } + return IO.NodeOutput(ui={"images": results}) -class GetImageSize: + save_svg = execute # TODO: remove + + +class GetImageSize(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return { - "required": { - "image": (IO.IMAGE,), - }, - "hidden": { - "unique_id": "UNIQUE_ID", - } - } + def define_schema(cls): + return IO.Schema( + node_id="GetImageSize", + display_name="Get Image Size", + description="Returns width and height of the image, and passes it through unchanged.", + category="image", + inputs=[ + IO.Image.Input("image"), + ], + outputs=[ + IO.Int.Output(display_name="width"), + IO.Int.Output(display_name="height"), + IO.Int.Output(display_name="batch_size"), + ], + hidden=[IO.Hidden.unique_id], + ) - RETURN_TYPES = (IO.INT, IO.INT, IO.INT) - RETURN_NAMES = ("width", "height", "batch_size") - FUNCTION = "get_size" - - CATEGORY = "image" - DESCRIPTION = """Returns width and height of the image, and passes it through unchanged.""" - - def get_size(self, image, unique_id=None) -> tuple[int, int]: + @classmethod + def execute(cls, image) -> IO.NodeOutput: height = image.shape[1] width = image.shape[2] batch_size = image.shape[0] # Send progress text to display size on the node - if unique_id: - PromptServer.instance.send_progress_text(f"width: {width}, height: {height}\n batch size: {batch_size}", unique_id) + if cls.hidden.unique_id: + PromptServer.instance.send_progress_text(f"width: {width}, height: {height}\n batch size: {batch_size}", cls.hidden.unique_id) - return width, height, batch_size + return IO.NodeOutput(width, height, batch_size) + + get_size = execute # TODO: remove + + +class ImageRotate(IO.ComfyNode): -class ImageRotate: @classmethod - def INPUT_TYPES(s): - return {"required": { "image": (IO.IMAGE,), - "rotation": (["none", "90 degrees", "180 degrees", "270 degrees"],), - }} - RETURN_TYPES = (IO.IMAGE,) - FUNCTION = "rotate" + def define_schema(cls): + return IO.Schema( + node_id="ImageRotate", + category="image/transform", + inputs=[ + IO.Image.Input("image"), + IO.Combo.Input("rotation", options=["none", "90 degrees", "180 degrees", "270 degrees"]), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image/transform" - - def rotate(self, image, rotation): + @classmethod + def execute(cls, image, rotation) -> IO.NodeOutput: rotate_by = 0 if rotation.startswith("90"): rotate_by = 1 @@ -604,41 +551,57 @@ class ImageRotate: rotate_by = 3 image = torch.rot90(image, k=rotate_by, dims=[2, 1]) - return (image,) + return IO.NodeOutput(image) + + rotate = execute # TODO: remove + + +class ImageFlip(IO.ComfyNode): -class ImageFlip: @classmethod - def INPUT_TYPES(s): - return {"required": { "image": (IO.IMAGE,), - "flip_method": (["x-axis: vertically", "y-axis: horizontally"],), - }} - RETURN_TYPES = (IO.IMAGE,) - FUNCTION = "flip" + def define_schema(cls): + return IO.Schema( + node_id="ImageFlip", + category="image/transform", + inputs=[ + IO.Image.Input("image"), + IO.Combo.Input("flip_method", options=["x-axis: vertically", "y-axis: horizontally"]), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image/transform" - - def flip(self, image, flip_method): + @classmethod + def execute(cls, image, flip_method) -> IO.NodeOutput: if flip_method.startswith("x"): image = torch.flip(image, dims=[1]) elif flip_method.startswith("y"): image = torch.flip(image, dims=[2]) - return (image,) + return IO.NodeOutput(image) -class ImageScaleToMaxDimension: - upscale_methods = ["area", "lanczos", "bilinear", "nearest-exact", "bilinear", "bicubic"] + flip = execute # TODO: remove + + +class ImageScaleToMaxDimension(IO.ComfyNode): @classmethod - def INPUT_TYPES(s): - return {"required": {"image": ("IMAGE",), - "upscale_method": (s.upscale_methods,), - "largest_size": ("INT", {"default": 512, "min": 0, "max": MAX_RESOLUTION, "step": 1})}} - RETURN_TYPES = ("IMAGE",) - FUNCTION = "upscale" + def define_schema(cls): + return IO.Schema( + node_id="ImageScaleToMaxDimension", + category="image/upscaling", + inputs=[ + IO.Image.Input("image"), + IO.Combo.Input( + "upscale_method", + options=["area", "lanczos", "bilinear", "nearest-exact", "bilinear", "bicubic"], + ), + IO.Int.Input("largest_size", default=512, min=0, max=MAX_RESOLUTION, step=1), + ], + outputs=[IO.Image.Output()], + ) - CATEGORY = "image/upscaling" - - def upscale(self, image, upscale_method, largest_size): + @classmethod + def execute(cls, image, upscale_method, largest_size) -> IO.NodeOutput: height = image.shape[1] width = image.shape[2] @@ -655,20 +618,30 @@ class ImageScaleToMaxDimension: samples = image.movedim(-1, 1) s = comfy.utils.common_upscale(samples, width, height, upscale_method, "disabled") s = s.movedim(1, -1) - return (s,) + return IO.NodeOutput(s) -NODE_CLASS_MAPPINGS = { - "ImageCrop": ImageCrop, - "RepeatImageBatch": RepeatImageBatch, - "ImageFromBatch": ImageFromBatch, - "ImageAddNoise": ImageAddNoise, - "SaveAnimatedWEBP": SaveAnimatedWEBP, - "SaveAnimatedPNG": SaveAnimatedPNG, - "SaveSVGNode": SaveSVGNode, - "ImageStitch": ImageStitch, - "ResizeAndPadImage": ResizeAndPadImage, - "GetImageSize": GetImageSize, - "ImageRotate": ImageRotate, - "ImageFlip": ImageFlip, - "ImageScaleToMaxDimension": ImageScaleToMaxDimension, -} + upscale = execute # TODO: remove + + +class ImagesExtension(ComfyExtension): + @override + async def get_node_list(self) -> list[type[IO.ComfyNode]]: + return [ + ImageCrop, + RepeatImageBatch, + ImageFromBatch, + ImageAddNoise, + SaveAnimatedWEBP, + SaveAnimatedPNG, + SaveSVGNode, + ImageStitch, + ResizeAndPadImage, + GetImageSize, + ImageRotate, + ImageFlip, + ImageScaleToMaxDimension, + ] + + +async def comfy_entrypoint() -> ImagesExtension: + return ImagesExtension() diff --git a/comfy_extras/nodes_qwen.py b/comfy_extras/nodes_qwen.py index 525239ae5..fde8fac9a 100644 --- a/comfy_extras/nodes_qwen.py +++ b/comfy_extras/nodes_qwen.py @@ -3,7 +3,9 @@ import comfy.utils import math from typing_extensions import override from comfy_api.latest import ComfyExtension, io - +import comfy.model_management +import torch +import nodes class TextEncodeQwenImageEdit(io.ComfyNode): @classmethod @@ -104,12 +106,37 @@ class TextEncodeQwenImageEditPlus(io.ComfyNode): return io.NodeOutput(conditioning) +class EmptyQwenImageLayeredLatentImage(io.ComfyNode): + @classmethod + def define_schema(cls): + return io.Schema( + node_id="EmptyQwenImageLayeredLatentImage", + display_name="Empty Qwen Image Layered Latent", + category="latent/qwen", + inputs=[ + io.Int.Input("width", default=640, min=16, max=nodes.MAX_RESOLUTION, step=16), + io.Int.Input("height", default=640, min=16, max=nodes.MAX_RESOLUTION, step=16), + io.Int.Input("layers", default=3, min=0, max=nodes.MAX_RESOLUTION, step=1), + io.Int.Input("batch_size", default=1, min=1, max=4096), + ], + outputs=[ + io.Latent.Output(), + ], + ) + + @classmethod + def execute(cls, width, height, layers, batch_size=1) -> io.NodeOutput: + latent = torch.zeros([batch_size, 16, layers + 1, height // 8, width // 8], device=comfy.model_management.intermediate_device()) + return io.NodeOutput({"samples": latent}) + + class QwenExtension(ComfyExtension): @override async def get_node_list(self) -> list[type[io.ComfyNode]]: return [ TextEncodeQwenImageEdit, TextEncodeQwenImageEditPlus, + EmptyQwenImageLayeredLatentImage, ] diff --git a/comfyui_version.py b/comfyui_version.py index b45309198..1f28e2407 100644 --- a/comfyui_version.py +++ b/comfyui_version.py @@ -1,3 +1,3 @@ # This file is automatically generated by the build process when version is # updated in pyproject.toml. -__version__ = "0.5.1" +__version__ = "0.6.0" diff --git a/manager_requirements.txt b/manager_requirements.txt index 2300f0c70..6585b0c19 100644 --- a/manager_requirements.txt +++ b/manager_requirements.txt @@ -1 +1 @@ -comfyui_manager==4.0.3b7 +comfyui_manager==4.0.4 diff --git a/pyproject.toml b/pyproject.toml index 3a6960811..35a268bd1 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,6 +1,6 @@ [project] name = "ComfyUI" -version = "0.5.1" +version = "0.6.0" readme = "README.md" license = { file = "LICENSE" } requires-python = ">=3.9" diff --git a/requirements.txt b/requirements.txt index 54696395f..8b670b813 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,5 +1,5 @@ -comfyui-frontend-package==1.34.9 -comfyui-workflow-templates==0.7.60 +comfyui-frontend-package==1.35.9 +comfyui-workflow-templates==0.7.64 comfyui-embedded-docs==0.3.1 torch torchsde diff --git a/tests-unit/comfy_extras_test/image_stitch_test.py b/tests-unit/comfy_extras_test/image_stitch_test.py index b5a0f022c..5c6a15ac4 100644 --- a/tests-unit/comfy_extras_test/image_stitch_test.py +++ b/tests-unit/comfy_extras_test/image_stitch_test.py @@ -25,7 +25,7 @@ class TestImageStitch: result = node.stitch(image1, "right", True, 0, "white", image2=None) - assert len(result) == 1 + assert len(result.result) == 1 assert torch.equal(result[0], image1) def test_basic_horizontal_stitch_right(self):