Merge branch 'comfyanonymous:master' into chroma-support

.github/workflows/update-api-stubs.yml

@@ -0,0 +1,47 @@
+name: Generate Pydantic Stubs from api.comfy.org
+
+on:
+  schedule:
+    - cron: '0 0 * * 1'
+  workflow_dispatch:
+
+jobs:
+  generate-models:
+    runs-on: ubuntu-latest
+    
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.10'
+      
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install 'datamodel-code-generator[http]'
+      
+      - name: Generate API models
+        run: |
+          datamodel-codegen  --use-subclass-enum --url https://api.comfy.org/openapi --output comfy_api_nodes/apis --output-model-type pydantic_v2.BaseModel
+      
+      - name: Check for changes
+        id: git-check
+        run: |
+          git diff --exit-code comfy_api_nodes/apis || echo "changes=true" >> $GITHUB_OUTPUT
+      
+      - name: Create Pull Request
+        if: steps.git-check.outputs.changes == 'true'
+        uses: peter-evans/create-pull-request@v5
+        with:
+          commit-message: 'chore: update API models from OpenAPI spec'
+          title: 'Update API models from api.comfy.org'
+          body: |
+            This PR updates the API models based on the latest api.comfy.org OpenAPI specification.
+            
+            Generated automatically by the a Github workflow.
+          branch: update-api-stubs
+          delete-branch: true
+          base: main

comfy/clip_vision.py

@@ -18,6 +18,7 @@ class Output:
         setattr(self, key, item)
 
 def clip_preprocess(image, size=224, mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711], crop=True):
+    image = image[:, :, :, :3] if image.shape[3] > 3 else image
     mean = torch.tensor(mean, device=image.device, dtype=image.dtype)
     std = torch.tensor(std, device=image.device, dtype=image.dtype)
     image = image.movedim(-1, 1)

comfy/comfy_types/node_typing.py

@@ -120,6 +120,10 @@ class InputTypeOptions(TypedDict):
     Available from frontend v1.17.5
     Ref: https://github.com/Comfy-Org/ComfyUI_frontend/pull/3548
     """
+    widgetType: NotRequired[str]
+    """Specifies a type to be used for widget initialization if different from the input type.
+    Available from frontend v1.18.0
+    https://github.com/Comfy-Org/ComfyUI_frontend/pull/3550"""
     # class InputTypeNumber(InputTypeOptions):
     # default: float | int
     min: NotRequired[float]

comfy/lora.py

@@ -279,6 +279,13 @@ def model_lora_keys_unet(model, key_map={}):
                 key_map["transformer.{}".format(key_lora)] = k
                 key_map["diffusion_model.{}".format(key_lora)] = k  # Old loras
 
+    if isinstance(model, comfy.model_base.HiDream):
+        for k in sdk:
+            if k.startswith("diffusion_model."):
+                if k.endswith(".weight"):
+                    key_lora = k[len("diffusion_model."):-len(".weight")].replace(".", "_")
+                    key_map["lycoris_{}".format(key_lora)] = k #SimpleTuner lycoris format
+
     return key_map
 
 

comfy/sd.py

@@ -121,6 +121,7 @@ class CLIP:
         self.layer_idx = None
         self.use_clip_schedule = False
         logging.info("CLIP/text encoder model load device: {}, offload device: {}, current: {}, dtype: {}".format(load_device, offload_device, params['device'], dtype))
+        self.tokenizer_options = {}
 
     def clone(self):
         n = CLIP(no_init=True)
@@ -128,6 +129,7 @@ class CLIP:
         n.cond_stage_model = self.cond_stage_model
         n.tokenizer = self.tokenizer
         n.layer_idx = self.layer_idx
+        n.tokenizer_options = self.tokenizer_options.copy()
         n.use_clip_schedule = self.use_clip_schedule
         n.apply_hooks_to_conds = self.apply_hooks_to_conds
         return n
@@ -135,10 +137,18 @@ class CLIP:
     def add_patches(self, patches, strength_patch=1.0, strength_model=1.0):
         return self.patcher.add_patches(patches, strength_patch, strength_model)
 
+    def set_tokenizer_option(self, option_name, value):
+        self.tokenizer_options[option_name] = value
+
     def clip_layer(self, layer_idx):
         self.layer_idx = layer_idx
 
     def tokenize(self, text, return_word_ids=False, **kwargs):
+        tokenizer_options = kwargs.get("tokenizer_options", {})
+        if len(self.tokenizer_options) > 0:
+            tokenizer_options = {**self.tokenizer_options, **tokenizer_options}
+        if len(tokenizer_options) > 0:
+            kwargs["tokenizer_options"] = tokenizer_options
         return self.tokenizer.tokenize_with_weights(text, return_word_ids, **kwargs)
 
     def add_hooks_to_dict(self, pooled_dict: dict[str]):

comfy/sd1_clip.py

@@ -457,13 +457,14 @@ def load_embed(embedding_name, embedding_directory, embedding_size, embed_key=No
     return embed_out
 
 class SDTokenizer:
-    def __init__(self, tokenizer_path=None, max_length=77, pad_with_end=True, embedding_directory=None, embedding_size=768, embedding_key='clip_l', tokenizer_class=CLIPTokenizer, has_start_token=True, has_end_token=True, pad_to_max_length=True, min_length=None, pad_token=None, end_token=None, tokenizer_data={}, tokenizer_args={}):
+    def __init__(self, tokenizer_path=None, max_length=77, pad_with_end=True, embedding_directory=None, embedding_size=768, embedding_key='clip_l', tokenizer_class=CLIPTokenizer, has_start_token=True, has_end_token=True, pad_to_max_length=True, min_length=None, pad_token=None, end_token=None, min_padding=None, tokenizer_data={}, tokenizer_args={}):
         if tokenizer_path is None:
             tokenizer_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), "sd1_tokenizer")
         self.tokenizer = tokenizer_class.from_pretrained(tokenizer_path, **tokenizer_args)
         self.max_length = tokenizer_data.get("{}_max_length".format(embedding_key), max_length)
         self.min_length = min_length
         self.end_token = None
+        self.min_padding = min_padding
 
         empty = self.tokenizer('')["input_ids"]
         self.tokenizer_adds_end_token = has_end_token
@@ -518,13 +519,15 @@ class SDTokenizer:
         return (embed, leftover)
 
 
-    def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
+    def tokenize_with_weights(self, text:str, return_word_ids=False, tokenizer_options={}, **kwargs):
         '''
         Takes a prompt and converts it to a list of (token, weight, word id) elements.
         Tokens can both be integer tokens and pre computed CLIP tensors.
         Word id values are unique per word and embedding, where the id 0 is reserved for non word tokens.
         Returned list has the dimensions NxM where M is the input size of CLIP
         '''
+        min_length = tokenizer_options.get("{}_min_length".format(self.embedding_key), self.min_length)
+        min_padding = tokenizer_options.get("{}_min_padding".format(self.embedding_key), self.min_padding)
 
         text = escape_important(text)
         parsed_weights = token_weights(text, 1.0)
@@ -603,10 +606,12 @@ class SDTokenizer:
         #fill last batch
         if self.end_token is not None:
             batch.append((self.end_token, 1.0, 0))
-        if self.pad_to_max_length:
+        if min_padding is not None:
+            batch.extend([(self.pad_token, 1.0, 0)] * min_padding)
+        if self.pad_to_max_length and len(batch) < self.max_length:
             batch.extend([(self.pad_token, 1.0, 0)] * (self.max_length - len(batch)))
-        if self.min_length is not None and len(batch) < self.min_length:
+        if min_length is not None and len(batch) < min_length:
-            batch.extend([(self.pad_token, 1.0, 0)] * (self.min_length - len(batch)))
+            batch.extend([(self.pad_token, 1.0, 0)] * (min_length - len(batch)))
 
         if not return_word_ids:
             batched_tokens = [[(t, w) for t, w,_ in x] for x in batched_tokens]
@@ -634,7 +639,7 @@ class SD1Tokenizer:
 
     def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
         out = {}
-        out[self.clip_name] = getattr(self, self.clip).tokenize_with_weights(text, return_word_ids)
+        out[self.clip_name] = getattr(self, self.clip).tokenize_with_weights(text, return_word_ids, **kwargs)
         return out
 
     def untokenize(self, token_weight_pair):

comfy/sdxl_clip.py

@@ -28,8 +28,8 @@ class SDXLTokenizer:
 
     def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
         out = {}
-        out["g"] = self.clip_g.tokenize_with_weights(text, return_word_ids)
+        out["g"] = self.clip_g.tokenize_with_weights(text, return_word_ids, **kwargs)
-        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids)
+        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids, **kwargs)
         return out
 
     def untokenize(self, token_weight_pair):

comfy/text_encoders/flux.py

@@ -19,8 +19,8 @@ class FluxTokenizer:
 
     def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
         out = {}
-        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids)
+        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids, **kwargs)
-        out["t5xxl"] = self.t5xxl.tokenize_with_weights(text, return_word_ids)
+        out["t5xxl"] = self.t5xxl.tokenize_with_weights(text, return_word_ids, **kwargs)
         return out
 
     def untokenize(self, token_weight_pair):

comfy/text_encoders/hidream.py

@@ -16,11 +16,11 @@ class HiDreamTokenizer:
 
     def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
         out = {}
-        out["g"] = self.clip_g.tokenize_with_weights(text, return_word_ids)
+        out["g"] = self.clip_g.tokenize_with_weights(text, return_word_ids, **kwargs)
-        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids)
+        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids, **kwargs)
-        t5xxl = self.t5xxl.tokenize_with_weights(text, return_word_ids)
+        t5xxl = self.t5xxl.tokenize_with_weights(text, return_word_ids, **kwargs)
         out["t5xxl"] = [t5xxl[0]]  # Use only first 128 tokens
-        out["llama"] = self.llama.tokenize_with_weights(text, return_word_ids)
+        out["llama"] = self.llama.tokenize_with_weights(text, return_word_ids, **kwargs)
         return out
 
     def untokenize(self, token_weight_pair):

comfy/text_encoders/hunyuan_video.py

@@ -49,13 +49,13 @@ class HunyuanVideoTokenizer:
 
     def tokenize_with_weights(self, text, return_word_ids=False, llama_template=None, image_embeds=None, image_interleave=1, **kwargs):
         out = {}
-        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids)
+        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids, **kwargs)
 
         if llama_template is None:
             llama_text = self.llama_template.format(text)
         else:
             llama_text = llama_template.format(text)
-        llama_text_tokens = self.llama.tokenize_with_weights(llama_text, return_word_ids)
+        llama_text_tokens = self.llama.tokenize_with_weights(llama_text, return_word_ids, **kwargs)
         embed_count = 0
         for r in llama_text_tokens:
             for i in range(len(r)):

comfy/text_encoders/hydit.py

@@ -41,8 +41,8 @@ class HyditTokenizer:
 
     def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
         out = {}
-        out["hydit_clip"] = self.hydit_clip.tokenize_with_weights(text, return_word_ids)
+        out["hydit_clip"] = self.hydit_clip.tokenize_with_weights(text, return_word_ids, **kwargs)
-        out["mt5xl"] = self.mt5xl.tokenize_with_weights(text, return_word_ids)
+        out["mt5xl"] = self.mt5xl.tokenize_with_weights(text, return_word_ids, **kwargs)
         return out
 
     def untokenize(self, token_weight_pair):

comfy/text_encoders/sd3_clip.py

@@ -45,9 +45,9 @@ class SD3Tokenizer:
 
     def tokenize_with_weights(self, text:str, return_word_ids=False, **kwargs):
         out = {}
-        out["g"] = self.clip_g.tokenize_with_weights(text, return_word_ids)
+        out["g"] = self.clip_g.tokenize_with_weights(text, return_word_ids, **kwargs)
-        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids)
+        out["l"] = self.clip_l.tokenize_with_weights(text, return_word_ids, **kwargs)
-        out["t5xxl"] = self.t5xxl.tokenize_with_weights(text, return_word_ids)
+        out["t5xxl"] = self.t5xxl.tokenize_with_weights(text, return_word_ids, **kwargs)
         return out
 
     def untokenize(self, token_weight_pair):

comfy_api_nodes/apis/PixverseController.py

@@ -0,0 +1,17 @@
+# generated by datamodel-codegen:
+#   filename:  https://api.comfy.org/openapi
+#   timestamp: 2025-04-23T15:56:33+00:00
+
+from __future__ import annotations
+
+from typing import Optional
+
+from pydantic import BaseModel
+
+from . import PixverseDto
+
+
+class ResponseData(BaseModel):
+    ErrCode: Optional[int] = None
+    ErrMsg: Optional[str] = None
+    Resp: Optional[PixverseDto.V2OpenAPII2VResp] = None

comfy_api_nodes/apis/PixverseDto.py

@@ -0,0 +1,57 @@
+# generated by datamodel-codegen:
+#   filename:  https://api.comfy.org/openapi
+#   timestamp: 2025-04-23T15:56:33+00:00
+
+from __future__ import annotations
+
+from typing import Optional
+
+from pydantic import BaseModel, Field, constr
+
+
+class V2OpenAPII2VResp(BaseModel):
+    video_id: Optional[int] = Field(None, description='Video_id')
+
+
+class V2OpenAPIT2VReq(BaseModel):
+    aspect_ratio: str = Field(
+        ..., description='Aspect ratio (16:9, 4:3, 1:1, 3:4, 9:16)', examples=['16:9']
+    )
+    duration: int = Field(
+        ...,
+        description='Video duration (5, 8 seconds, --model=v3.5 only allows 5,8; --quality=1080p does not support 8s)',
+        examples=[5],
+    )
+    model: str = Field(
+        ..., description='Model version (only supports v3.5)', examples=['v3.5']
+    )
+    motion_mode: Optional[str] = Field(
+        'normal',
+        description='Motion mode (normal, fast, --fast only available when duration=5; --quality=1080p does not support fast)',
+        examples=['normal'],
+    )
+    negative_prompt: Optional[constr(max_length=2048)] = Field(
+        None, description='Negative prompt\n'
+    )
+    prompt: constr(max_length=2048) = Field(..., description='Prompt')
+    quality: str = Field(
+        ...,
+        description='Video quality ("360p"(Turbo model), "540p", "720p", "1080p")',
+        examples=['540p'],
+    )
+    seed: Optional[int] = Field(None, description='Random seed, range: 0 - 2147483647')
+    style: Optional[str] = Field(
+        None,
+        description='Style (effective when model=v3.5, "anime", "3d_animation", "clay", "comic", "cyberpunk") Do not include style parameter unless needed',
+        examples=['anime'],
+    )
+    template_id: Optional[int] = Field(
+        None,
+        description='Template ID (template_id must be activated before use)',
+        examples=[302325299692608],
+    )
+    water_mark: Optional[bool] = Field(
+        False,
+        description='Watermark (true: add watermark, false: no watermark)',
+        examples=[False],
+    )

comfy_api_nodes/apis/__init__.py

@@ -0,0 +1,422 @@
+# generated by datamodel-codegen:
+#   filename:  https://api.comfy.org/openapi
+#   timestamp: 2025-04-23T15:56:33+00:00
+
+from __future__ import annotations
+
+from datetime import datetime
+from enum import Enum
+from typing import Any, Dict, List, Optional
+
+from pydantic import AnyUrl, BaseModel, Field, confloat, conint
+
+class Customer(BaseModel):
+    createdAt: Optional[datetime] = Field(
+        None, description='The date and time the user was created'
+    )
+    email: Optional[str] = Field(None, description='The email address for this user')
+    id: str = Field(..., description='The firebase UID of the user')
+    name: Optional[str] = Field(None, description='The name for this user')
+    updatedAt: Optional[datetime] = Field(
+        None, description='The date and time the user was last updated'
+    )
+
+
+class Error(BaseModel):
+    details: Optional[List[str]] = Field(
+        None,
+        description='Optional detailed information about the error or hints for resolving it.',
+    )
+    message: Optional[str] = Field(
+        None, description='A clear and concise description of the error.'
+    )
+
+
+class ErrorResponse(BaseModel):
+    error: str
+    message: str
+
+class ImageRequest(BaseModel):
+    aspect_ratio: Optional[str] = Field(
+        None,
+        description="Optional. The aspect ratio (e.g., 'ASPECT_16_9', 'ASPECT_1_1'). Cannot be used with resolution. Defaults to 'ASPECT_1_1' if unspecified.",
+    )
+    color_palette: Optional[Dict[str, Any]] = Field(
+        None, description='Optional. Color palette object. Only for V_2, V_2_TURBO.'
+    )
+    magic_prompt_option: Optional[str] = Field(
+        None, description="Optional. MagicPrompt usage ('AUTO', 'ON', 'OFF')."
+    )
+    model: str = Field(..., description="The model used (e.g., 'V_2', 'V_2A_TURBO')")
+    negative_prompt: Optional[str] = Field(
+        None,
+        description='Optional. Description of what to exclude. Only for V_1, V_1_TURBO, V_2, V_2_TURBO.',
+    )
+    num_images: Optional[conint(ge=1, le=8)] = Field(
+        1, description='Optional. Number of images to generate (1-8). Defaults to 1.'
+    )
+    prompt: str = Field(
+        ..., description='Required. The prompt to use to generate the image.'
+    )
+    resolution: Optional[str] = Field(
+        None,
+        description="Optional. Resolution (e.g., 'RESOLUTION_1024_1024'). Only for model V_2. Cannot be used with aspect_ratio.",
+    )
+    seed: Optional[conint(ge=0, le=2147483647)] = Field(
+        None, description='Optional. A number between 0 and 2147483647.'
+    )
+    style_type: Optional[str] = Field(
+        None,
+        description="Optional. Style type ('AUTO', 'GENERAL', 'REALISTIC', 'DESIGN', 'RENDER_3D', 'ANIME'). Only for models V_2 and above.",
+    )
+
+
+class Datum(BaseModel):
+    is_image_safe: Optional[bool] = Field(
+        None, description='Indicates whether the image is considered safe.'
+    )
+    prompt: Optional[str] = Field(
+        None, description='The prompt used to generate this image.'
+    )
+    resolution: Optional[str] = Field(
+        None, description="The resolution of the generated image (e.g., '1024x1024')."
+    )
+    seed: Optional[int] = Field(
+        None, description='The seed value used for this generation.'
+    )
+    style_type: Optional[str] = Field(
+        None,
+        description="The style type used for generation (e.g., 'REALISTIC', 'ANIME').",
+    )
+    url: Optional[str] = Field(None, description='URL to the generated image.')
+
+
+class Code(Enum):
+    int_1100 = 1100
+    int_1101 = 1101
+    int_1102 = 1102
+    int_1103 = 1103
+
+
+class Code1(Enum):
+    int_1000 = 1000
+    int_1001 = 1001
+    int_1002 = 1002
+    int_1003 = 1003
+    int_1004 = 1004
+
+
+class AspectRatio(str, Enum):
+    field_16_9 = '16:9'
+    field_9_16 = '9:16'
+    field_1_1 = '1:1'
+
+
+class Config(BaseModel):
+    horizontal: Optional[confloat(ge=-10.0, le=10.0)] = None
+    pan: Optional[confloat(ge=-10.0, le=10.0)] = None
+    roll: Optional[confloat(ge=-10.0, le=10.0)] = None
+    tilt: Optional[confloat(ge=-10.0, le=10.0)] = None
+    vertical: Optional[confloat(ge=-10.0, le=10.0)] = None
+    zoom: Optional[confloat(ge=-10.0, le=10.0)] = None
+
+
+class Type(str, Enum):
+    simple = 'simple'
+    down_back = 'down_back'
+    forward_up = 'forward_up'
+    right_turn_forward = 'right_turn_forward'
+    left_turn_forward = 'left_turn_forward'
+
+
+class CameraControl(BaseModel):
+    config: Optional[Config] = None
+    type: Optional[Type] = Field(None, description='Predefined camera movements type')
+
+
+class Duration(str, Enum):
+    field_5 = 5
+    field_10 = 10
+
+
+class Mode(str, Enum):
+    std = 'std'
+    pro = 'pro'
+
+
+class TaskInfo(BaseModel):
+    external_task_id: Optional[str] = None
+
+
+class Video(BaseModel):
+    duration: Optional[str] = Field(None, description='Total video duration')
+    id: Optional[str] = Field(None, description='Generated video ID')
+    url: Optional[AnyUrl] = Field(None, description='URL for generated video')
+
+
+class TaskResult(BaseModel):
+    videos: Optional[List[Video]] = None
+
+
+class TaskStatus(str, Enum):
+    submitted = 'submitted'
+    processing = 'processing'
+    succeed = 'succeed'
+    failed = 'failed'
+
+
+class Data(BaseModel):
+    created_at: Optional[int] = Field(None, description='Task creation time')
+    task_id: Optional[str] = Field(None, description='Task ID')
+    task_info: Optional[TaskInfo] = None
+    task_result: Optional[TaskResult] = None
+    task_status: Optional[TaskStatus] = None
+    updated_at: Optional[int] = Field(None, description='Task update time')
+
+
+class AspectRatio1(str, Enum):
+    field_16_9 = '16:9'
+    field_9_16 = '9:16'
+    field_1_1 = '1:1'
+    field_4_3 = '4:3'
+    field_3_4 = '3:4'
+    field_3_2 = '3:2'
+    field_2_3 = '2:3'
+    field_21_9 = '21:9'
+
+
+class ImageReference(str, Enum):
+    subject = 'subject'
+    face = 'face'
+
+
+class Image(BaseModel):
+    index: Optional[int] = Field(None, description='Image Number (0-9)')
+    url: Optional[AnyUrl] = Field(None, description='URL for generated image')
+
+
+class TaskResult1(BaseModel):
+    images: Optional[List[Image]] = None
+
+
+class Data1(BaseModel):
+    created_at: Optional[int] = Field(None, description='Task creation time')
+    task_id: Optional[str] = Field(None, description='Task ID')
+    task_result: Optional[TaskResult1] = None
+    task_status: Optional[TaskStatus] = None
+    task_status_msg: Optional[str] = Field(None, description='Task status information')
+    updated_at: Optional[int] = Field(None, description='Task update time')
+
+
+class AspectRatio2(str, Enum):
+    field_16_9 = '16:9'
+    field_9_16 = '9:16'
+    field_1_1 = '1:1'
+
+
+class CameraControl1(BaseModel):
+    config: Optional[Config] = None
+    type: Optional[Type] = Field(None, description='Predefined camera movements type')
+
+
+class ModelName2(str, Enum):
+    kling_v1 = 'kling-v1'
+    kling_v1_6 = 'kling-v1-6'
+
+
+class TaskResult2(BaseModel):
+    videos: Optional[List[Video]] = None
+
+
+class Data2(BaseModel):
+    created_at: Optional[int] = Field(None, description='Task creation time')
+    task_id: Optional[str] = Field(None, description='Task ID')
+    task_info: Optional[TaskInfo] = None
+    task_result: Optional[TaskResult2] = None
+    task_status: Optional[TaskStatus] = None
+    updated_at: Optional[int] = Field(None, description='Task update time')
+
+
+class Code2(Enum):
+    int_1200 = 1200
+    int_1201 = 1201
+    int_1202 = 1202
+    int_1203 = 1203
+
+
+class ResourcePackType(str, Enum):
+    decreasing_total = 'decreasing_total'
+    constant_period = 'constant_period'
+
+
+class Status(str, Enum):
+    toBeOnline = 'toBeOnline'
+    online = 'online'
+    expired = 'expired'
+    runOut = 'runOut'
+
+
+class ResourcePackSubscribeInfo(BaseModel):
+    effective_time: Optional[int] = Field(
+        None, description='Effective time, Unix timestamp in ms'
+    )
+    invalid_time: Optional[int] = Field(
+        None, description='Expiration time, Unix timestamp in ms'
+    )
+    purchase_time: Optional[int] = Field(
+        None, description='Purchase time, Unix timestamp in ms'
+    )
+    remaining_quantity: Optional[float] = Field(
+        None, description='Remaining quantity (updated with a 12-hour delay)'
+    )
+    resource_pack_id: Optional[str] = Field(None, description='Resource package ID')
+    resource_pack_name: Optional[str] = Field(None, description='Resource package name')
+    resource_pack_type: Optional[ResourcePackType] = Field(
+        None,
+        description='Resource package type (decreasing_total=decreasing total, constant_period=constant periodicity)',
+    )
+    status: Optional[Status] = Field(None, description='Resource Package Status')
+    total_quantity: Optional[float] = Field(None, description='Total quantity')
+
+class Background(str, Enum):
+    transparent = 'transparent'
+    opaque = 'opaque'
+
+
+class Moderation(str, Enum):
+    low = 'low'
+    auto = 'auto'
+
+
+class OutputFormat(str, Enum):
+    png = 'png'
+    webp = 'webp'
+    jpeg = 'jpeg'
+
+
+class Quality(str, Enum):
+    low = 'low'
+    medium = 'medium'
+    high = 'high'
+
+
+class OpenAIImageEditRequest(BaseModel):
+    background: Optional[str] = Field(
+        None, description='Background transparency', examples=['opaque']
+    )
+    model: str = Field(
+        ..., description='The model to use for image editing', examples=['gpt-image-1']
+    )
+    moderation: Optional[Moderation] = Field(
+        None, description='Content moderation setting', examples=['auto']
+    )
+    n: Optional[int] = Field(
+        None, description='The number of images to generate', examples=[1]
+    )
+    output_compression: Optional[int] = Field(
+        None, description='Compression level for JPEG or WebP (0-100)', examples=[100]
+    )
+    output_format: Optional[OutputFormat] = Field(
+        None, description='Format of the output image', examples=['png']
+    )
+    prompt: str = Field(
+        ...,
+        description='A text description of the desired edit',
+        examples=['Give the rocketship rainbow coloring'],
+    )
+    quality: Optional[str] = Field(
+        None, description='The quality of the edited image', examples=['low']
+    )
+    size: Optional[str] = Field(
+        None, description='Size of the output image', examples=['1024x1024']
+    )
+    user: Optional[str] = Field(
+        None,
+        description='A unique identifier for end-user monitoring',
+        examples=['user-1234'],
+    )
+
+
+class Quality1(str, Enum):
+    low = 'low'
+    medium = 'medium'
+    high = 'high'
+    standard = 'standard'
+    hd = 'hd'
+
+
+class ResponseFormat(str, Enum):
+    url = 'url'
+    b64_json = 'b64_json'
+
+
+class Style(str, Enum):
+    vivid = 'vivid'
+    natural = 'natural'
+
+
+class OpenAIImageGenerationRequest(BaseModel):
+    background: Optional[Background] = Field(
+        None, description='Background transparency', examples=['opaque']
+    )
+    model: Optional[str] = Field(
+        None, description='The model to use for image generation', examples=['dall-e-3']
+    )
+    moderation: Optional[Moderation] = Field(
+        None, description='Content moderation setting', examples=['auto']
+    )
+    n: Optional[int] = Field(
+        None,
+        description='The number of images to generate (1-10). Only 1 supported for dall-e-3.',
+        examples=[1],
+    )
+    output_compression: Optional[int] = Field(
+        None, description='Compression level for JPEG or WebP (0-100)', examples=[100]
+    )
+    output_format: Optional[OutputFormat] = Field(
+        None, description='Format of the output image', examples=['png']
+    )
+    prompt: str = Field(
+        ...,
+        description='A text description of the desired image',
+        examples=['Draw a rocket in front of a blackhole in deep space'],
+    )
+    quality: Optional[Quality1] = Field(
+        None, description='The quality of the generated image', examples=['high']
+    )
+    response_format: Optional[ResponseFormat] = Field(
+        None, description='Response format of image data', examples=['b64_json']
+    )
+    size: Optional[str] = Field(
+        None,
+        description='Size of the image (e.g., 1024x1024, 1536x1024, auto)',
+        examples=['1024x1536'],
+    )
+    style: Optional[Style] = Field(
+        None, description='Style of the image (only for dall-e-3)', examples=['vivid']
+    )
+    user: Optional[str] = Field(
+        None,
+        description='A unique identifier for end-user monitoring',
+        examples=['user-1234'],
+    )
+
+
+class Datum1(BaseModel):
+    b64_json: Optional[str] = Field(None, description='Base64 encoded image data')
+    revised_prompt: Optional[str] = Field(None, description='Revised prompt')
+    url: Optional[str] = Field(None, description='URL of the image')
+
+
+class OpenAIImageGenerationResponse(BaseModel):
+    data: Optional[List[Datum1]] = None
+class User(BaseModel):
+    email: Optional[str] = Field(None, description='The email address for this user.')
+    id: Optional[str] = Field(None, description='The unique id for this user.')
+    isAdmin: Optional[bool] = Field(
+        None, description='Indicates if the user has admin privileges.'
+    )
+    isApproved: Optional[bool] = Field(
+        None, description='Indicates if the user is approved.'
+    )
+    name: Optional[str] = Field(None, description='The name for this user.')

comfy_api_nodes/apis/client.py

@@ -226,7 +226,7 @@ class ApiClient:
     def check_auth_token(self, auth_token):
         """Verify that an auth token is present."""
         if auth_token is None:
-            raise Exception("Please login first to use this node.")
+            raise Exception("Unauthorized: Please login first to use this node.")
         return auth_token
 
 
@@ -269,7 +269,7 @@ class SynchronousOperation(Generic[T, R]):
         files: Optional[Dict[str, Any]] = None,
         api_base: str = "https://api.comfy.org",
         auth_token: Optional[str] = None,
-        timeout: float = 60.0,
+        timeout: float = 604800.0,
         verify_ssl: bool = True,
     ):
         self.endpoint = endpoint

comfy_api_nodes/nodes_api.py

@@ -0,0 +1,442 @@
+import io
+from inspect import cleandoc
+
+from comfy.utils import common_upscale
+from comfy.comfy_types.node_typing import IO, ComfyNodeABC, InputTypeDict
+from comfy_api_nodes.apis import (
+    OpenAIImageGenerationRequest,
+    OpenAIImageEditRequest,
+    OpenAIImageGenerationResponse
+)
+from comfy_api_nodes.apis.client import ApiEndpoint, HttpMethod, SynchronousOperation
+
+import numpy as np
+from PIL import Image
+import requests
+import torch
+import math
+import base64
+
+def downscale_input(image):
+    samples = image.movedim(-1,1)
+    #downscaling input images to roughly the same size as the outputs
+    total = int(1536 * 1024)
+    scale_by = math.sqrt(total / (samples.shape[3] * samples.shape[2]))
+    if scale_by >= 1:
+        return image
+    width = round(samples.shape[3] * scale_by)
+    height = round(samples.shape[2] * scale_by)
+
+    s = common_upscale(samples, width, height, "lanczos", "disabled")
+    s = s.movedim(1,-1)
+    return s
+
+def validate_and_cast_response(response):
+    # validate raw JSON response
+    data = response.data
+    if not data or len(data) == 0:
+        raise Exception("No images returned from API endpoint")
+
+    # Initialize list to store image tensors
+    image_tensors = []
+
+    # Process each image in the data array
+    for image_data in data:
+        image_url = image_data.url
+        b64_data = image_data.b64_json
+
+        if not image_url and not b64_data:
+            raise Exception("No image was generated in the response")
+
+        if b64_data:
+            img_data = base64.b64decode(b64_data)
+            img = Image.open(io.BytesIO(img_data))
+
+        elif image_url:
+            img_response = requests.get(image_url)
+            if img_response.status_code != 200:
+                raise Exception("Failed to download the image")
+            img = Image.open(io.BytesIO(img_response.content))
+
+        img = img.convert("RGBA")
+
+        # Convert to numpy array, normalize to float32 between 0 and 1
+        img_array = np.array(img).astype(np.float32) / 255.0
+        img_tensor = torch.from_numpy(img_array)
+
+        # Add to list of tensors
+        image_tensors.append(img_tensor)
+
+    return torch.stack(image_tensors, dim=0)
+
+class OpenAIDalle2(ComfyNodeABC):
+    """
+    Generates images synchronously via OpenAI's DALL·E 2 endpoint.
+
+    Uses the proxy at /proxy/openai/images/generations. Returned URLs are short‑lived,
+    so download or cache results if you need to keep them.
+    """
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls) -> InputTypeDict:
+        return {
+            "required": {
+                "prompt": (IO.STRING, {
+                    "multiline": True,
+                    "default": "",
+                    "tooltip": "Text prompt for DALL·E",
+                }),
+            },
+            "optional": {
+                "seed": (IO.INT, {
+                    "default": 0,
+                    "min": 0,
+                    "max": 2**31-1,
+                    "step": 1,
+                    "display": "number",
+                    "tooltip": "not implemented yet in backend",
+                }),
+                "size": (IO.COMBO, {
+                    "options": ["256x256", "512x512", "1024x1024"],
+                    "default": "1024x1024",
+                    "tooltip": "Image size",
+                }),
+                "n": (IO.INT, {
+                    "default": 1,
+                    "min": 1,
+                    "max": 8,
+                    "step": 1,
+                    "display": "number",
+                    "tooltip": "How many images to generate",
+                }),
+                "image": (IO.IMAGE, {
+                    "default": None,
+                    "tooltip": "Optional reference image for image editing.",
+                }),
+                "mask": (IO.MASK, {
+                    "default": None,
+                    "tooltip": "Optional mask for inpainting (white areas will be replaced)",
+                }),
+            },
+            "hidden": {
+                "auth_token": "AUTH_TOKEN_COMFY_ORG"
+            }
+        }
+
+    RETURN_TYPES = (IO.IMAGE,)
+    FUNCTION = "api_call"
+    CATEGORY = "api node"
+    DESCRIPTION = cleandoc(__doc__ or "")
+    API_NODE = True
+
+    def api_call(self, prompt, seed=0, image=None, mask=None, n=1, size="1024x1024", auth_token=None):
+        model = "dall-e-2"
+        path = "/proxy/openai/images/generations"
+        request_class = OpenAIImageGenerationRequest
+        img_binary = None
+
+        if image is not None and mask is not None:
+            path = "/proxy/openai/images/edits"
+            request_class = OpenAIImageEditRequest
+
+            input_tensor = image.squeeze().cpu()
+            height, width, channels = input_tensor.shape
+            rgba_tensor = torch.ones(height, width, 4, device="cpu")
+            rgba_tensor[:, :, :channels] = input_tensor
+
+            if mask.shape[1:] != image.shape[1:-1]:
+                raise Exception("Mask and Image must be the same size")
+            rgba_tensor[:,:,3] = (1-mask.squeeze().cpu())
+
+            rgba_tensor = downscale_input(rgba_tensor.unsqueeze(0)).squeeze()
+
+            image_np = (rgba_tensor.numpy() * 255).astype(np.uint8)
+            img = Image.fromarray(image_np)
+            img_byte_arr = io.BytesIO()
+            img.save(img_byte_arr, format='PNG')
+            img_byte_arr.seek(0)
+            img_binary = img_byte_arr#.getvalue()
+            img_binary.name = "image.png"
+        elif image is not None or mask is not None:
+            raise Exception("Dall-E 2 image editing requires an image AND a mask")
+
+        # Build the operation
+        operation = SynchronousOperation(
+            endpoint=ApiEndpoint(
+                path=path,
+                method=HttpMethod.POST,
+                request_model=request_class,
+                response_model=OpenAIImageGenerationResponse
+            ),
+            request=request_class(
+                model=model,
+                prompt=prompt,
+                n=n,
+                size=size,
+                seed=seed,
+            ),
+            files={
+                "image": img_binary,
+            } if img_binary else None,
+            auth_token=auth_token
+        )
+
+        response = operation.execute()
+
+        img_tensor = validate_and_cast_response(response)
+        return (img_tensor,)
+
+class OpenAIDalle3(ComfyNodeABC):
+    """
+    Generates images synchronously via OpenAI's DALL·E 3 endpoint.
+
+    Uses the proxy at /proxy/openai/images/generations. Returned URLs are short‑lived,
+    so download or cache results if you need to keep them.
+    """
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls) -> InputTypeDict:
+        return {
+            "required": {
+                "prompt": (IO.STRING, {
+                    "multiline": True,
+                    "default": "",
+                    "tooltip": "Text prompt for DALL·E",
+                }),
+            },
+            "optional": {
+                "seed": (IO.INT, {
+                    "default": 0,
+                    "min": 0,
+                    "max": 2**31-1,
+                    "step": 1,
+                    "display": "number",
+                    "tooltip": "not implemented yet in backend",
+                }),
+                "quality" : (IO.COMBO, {
+                    "options": ["standard","hd"],
+                    "default": "standard",
+                    "tooltip": "Image quality",
+                }),
+                "style": (IO.COMBO, {
+                    "options": ["natural","vivid"],
+                    "default": "natural",
+                    "tooltip": "Vivid causes the model to lean towards generating hyper-real and dramatic images. Natural causes the model to produce more natural, less hyper-real looking images.",
+                }),
+                "size": (IO.COMBO, {
+                    "options": ["1024x1024", "1024x1792", "1792x1024"],
+                    "default": "1024x1024",
+                    "tooltip": "Image size",
+                }),
+            },
+            "hidden": {
+                "auth_token": "AUTH_TOKEN_COMFY_ORG"
+            }
+        }
+
+    RETURN_TYPES = (IO.IMAGE,)
+    FUNCTION = "api_call"
+    CATEGORY = "api node"
+    DESCRIPTION = cleandoc(__doc__ or "")
+    API_NODE = True
+
+    def api_call(self, prompt, seed=0, style="natural", quality="standard", size="1024x1024", auth_token=None):
+        model = "dall-e-3"
+
+        # build the operation
+        operation = SynchronousOperation(
+            endpoint=ApiEndpoint(
+                path="/proxy/openai/images/generations",
+                method=HttpMethod.POST,
+                request_model=OpenAIImageGenerationRequest,
+                response_model=OpenAIImageGenerationResponse
+            ),
+            request=OpenAIImageGenerationRequest(
+                model=model,
+                prompt=prompt,
+                quality=quality,
+                size=size,
+                style=style,
+                seed=seed,
+            ),
+            auth_token=auth_token
+        )
+
+        response = operation.execute()
+
+        img_tensor = validate_and_cast_response(response)
+        return (img_tensor,)
+
+class OpenAIGPTImage1(ComfyNodeABC):
+    """
+    Generates images synchronously via OpenAI's GPT Image 1 endpoint.
+
+    Uses the proxy at /proxy/openai/images/generations. Returned URLs are short‑lived,
+    so download or cache results if you need to keep them.
+    """
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls) -> InputTypeDict:
+        return {
+            "required": {
+                "prompt": (IO.STRING, {
+                    "multiline": True,
+                    "default": "",
+                    "tooltip": "Text prompt for GPT Image 1",
+                }),
+            },
+            "optional": {
+                "seed": (IO.INT, {
+                    "default": 0,
+                    "min": 0,
+                    "max": 2**31-1,
+                    "step": 1,
+                    "display": "number",
+                    "tooltip": "not implemented yet in backend",
+                }),
+                "quality": (IO.COMBO, {
+                    "options": ["low","medium","high"],
+                    "default": "low",
+                    "tooltip": "Image quality, affects cost and generation time.",
+                }),
+                "background": (IO.COMBO, {
+                    "options": ["opaque","transparent"],
+                    "default": "opaque",
+                    "tooltip": "Return image with or without background",
+                }),
+                "size": (IO.COMBO, {
+                    "options": ["auto", "1024x1024", "1024x1536", "1536x1024"],
+                    "default": "auto",
+                    "tooltip": "Image size",
+                }),
+                "n": (IO.INT, {
+                    "default": 1,
+                    "min": 1,
+                    "max": 8,
+                    "step": 1,
+                    "display": "number",
+                    "tooltip": "How many images to generate",
+                }),
+                "image": (IO.IMAGE, {
+                    "default": None,
+                    "tooltip": "Optional reference image for image editing.",
+                }),
+                "mask": (IO.MASK, {
+                    "default": None,
+                    "tooltip": "Optional mask for inpainting (white areas will be replaced)",
+                }),
+            },
+            "hidden": {
+                "auth_token": "AUTH_TOKEN_COMFY_ORG"
+            }
+        }
+
+    RETURN_TYPES = (IO.IMAGE,)
+    FUNCTION = "api_call"
+    CATEGORY = "api node"
+    DESCRIPTION = cleandoc(__doc__ or "")
+    API_NODE = True
+
+    def api_call(self, prompt, seed=0, quality="low", background="opaque", image=None, mask=None, n=1, size="1024x1024", auth_token=None):
+        model = "gpt-image-1"
+        path = "/proxy/openai/images/generations"
+        request_class = OpenAIImageGenerationRequest
+        img_binaries = []
+        mask_binary = None
+        files = []
+
+        if image is not None:
+            path = "/proxy/openai/images/edits"
+            request_class = OpenAIImageEditRequest
+
+            batch_size = image.shape[0]
+
+
+            for i in range(batch_size):
+                single_image = image[i:i+1]
+                scaled_image = downscale_input(single_image).squeeze()
+
+                image_np = (scaled_image.numpy() * 255).astype(np.uint8)
+                img = Image.fromarray(image_np)
+                img_byte_arr = io.BytesIO()
+                img.save(img_byte_arr, format='PNG')
+                img_byte_arr.seek(0)
+                img_binary = img_byte_arr
+                img_binary.name = f"image_{i}.png"
+
+                img_binaries.append(img_binary)
+                if batch_size == 1:
+                    files.append(("image", img_binary))
+                else:
+                    files.append(("image[]", img_binary))
+
+        if mask is not None:
+            if image.shape[0] != 1:
+                raise Exception("Cannot use a mask with multiple image")
+            if image is None:
+                raise Exception("Cannot use a mask without an input image")
+            if mask.shape[1:] != image.shape[1:-1]:
+                raise Exception("Mask and Image must be the same size")
+            batch, height, width = mask.shape
+            rgba_mask = torch.zeros(height, width, 4, device="cpu")
+            rgba_mask[:,:,3] = (1-mask.squeeze().cpu())
+
+            scaled_mask = downscale_input(rgba_mask.unsqueeze(0)).squeeze()
+
+            mask_np = (scaled_mask.numpy() * 255).astype(np.uint8)
+            mask_img = Image.fromarray(mask_np)
+            mask_img_byte_arr = io.BytesIO()
+            mask_img.save(mask_img_byte_arr, format='PNG')
+            mask_img_byte_arr.seek(0)
+            mask_binary = mask_img_byte_arr
+            mask_binary.name = "mask.png"
+            files.append(("mask", mask_binary))
+
+
+        # Build the operation
+        operation = SynchronousOperation(
+            endpoint=ApiEndpoint(
+                path=path,
+                method=HttpMethod.POST,
+                request_model=request_class,
+                response_model=OpenAIImageGenerationResponse
+            ),
+            request=request_class(
+                model=model,
+                prompt=prompt,
+                quality=quality,
+                background=background,
+                n=n,
+                seed=seed,
+                size=size,
+            ),
+            files=files if files else None,
+            auth_token=auth_token
+        )
+
+        response = operation.execute()
+
+        img_tensor = validate_and_cast_response(response)
+        return (img_tensor,)
+
+
+# A dictionary that contains all nodes you want to export with their names
+# NOTE: names should be globally unique
+NODE_CLASS_MAPPINGS = {
+    "OpenAIDalle2": OpenAIDalle2,
+    "OpenAIDalle3": OpenAIDalle3,
+    "OpenAIGPTImage1": OpenAIGPTImage1,
+}
+
+# A dictionary that contains the friendly/humanly readable titles for the nodes
+NODE_DISPLAY_NAME_MAPPINGS = {
+    "OpenAIDalle2": "OpenAI DALL·E 2",
+    "OpenAIDalle3": "OpenAI DALL·E 3",
+    "OpenAIGPTImage1": "OpenAI GPT Image 1",
+}

comfy_extras/nodes_cond.py

@@ -20,6 +20,29 @@ class CLIPTextEncodeControlnet:
             c.append(n)
         return (c, )
 
+class T5TokenizerOptions:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+            "required": {
+                "clip": ("CLIP", ),
+                "min_padding": ("INT", {"default": 0, "min": 0, "max": 10000, "step": 1}),
+                "min_length": ("INT", {"default": 0, "min": 0, "max": 10000, "step": 1}),
+            }
+        }
+
+    RETURN_TYPES = ("CLIP",)
+    FUNCTION = "set_options"
+
+    def set_options(self, clip, min_padding, min_length):
+        clip = clip.clone()
+        for t5_type in ["t5xxl", "pile_t5xl", "t5base", "mt5xl", "umt5xxl"]:
+            clip.set_tokenizer_option("{}_min_padding".format(t5_type), min_padding)
+            clip.set_tokenizer_option("{}_min_length".format(t5_type), min_length)
+
+        return (clip, )
+
 NODE_CLASS_MAPPINGS = {
-    "CLIPTextEncodeControlnet": CLIPTextEncodeControlnet
+    "CLIPTextEncodeControlnet": CLIPTextEncodeControlnet,
+    "T5TokenizerOptions": T5TokenizerOptions,
 }

comfy_extras/nodes_lt.py

@@ -385,7 +385,7 @@ def encode_single_frame(output_file, image_array: np.ndarray, crf):
     container = av.open(output_file, "w", format="mp4")
     try:
         stream = container.add_stream(
-            "h264", rate=1, options={"crf": str(crf), "preset": "veryfast"}
+            "libx264", rate=1, options={"crf": str(crf), "preset": "veryfast"}
         )
         stream.height = image_array.shape[0]
         stream.width = image_array.shape[1]

comfyui_version.py

@@ -1,3 +1,3 @@
 # This file is automatically generated by the build process when version is
 # updated in pyproject.toml.
-__version__ = "0.3.29"
+__version__ = "0.3.30"

nodes.py

@@ -2260,11 +2260,20 @@ def init_builtin_extra_nodes():
         "nodes_fresca.py",
     ]
 
+    api_nodes_dir = os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_api_nodes")
+    api_nodes_files = [
+        "nodes_api.py",
+    ]
+
     import_failed = []
     for node_file in extras_files:
         if not load_custom_node(os.path.join(extras_dir, node_file), module_parent="comfy_extras"):
             import_failed.append(node_file)
 
+    for node_file in api_nodes_files:
+        if not load_custom_node(os.path.join(api_nodes_dir, node_file), module_parent="comfy_api_nodes"):
+            import_failed.append(node_file)
+
     return import_failed
 
 

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ComfyUI"
-version = "0.3.29"
+version = "0.3.30"
 readme = "README.md"
 license = { file = "LICENSE" }
 requires-python = ">=3.9"

requirements.txt

@@ -1,4 +1,4 @@
-comfyui-frontend-package==1.16.9
+comfyui-frontend-package==1.17.11
 comfyui-workflow-templates==0.1.3
 torch
 torchsde