Merge branch 'master' into deepme987/add-execution-environment-api

Adds api.environment.get() to the public ComfyAPI — returns
"local" (default), "cloud", or "remote" based on the
COMFY_EXECUTION_ENVIRONMENT env var.

Custom nodes use this to adapt behavior for managed deployments
(e.g. skip model downloads when models are pre-provisioned).

comfy/model_base.py

@@ -21,7 +21,6 @@ import comfy.ldm.hunyuan3dv2_1.hunyuandit
 import torch
 import logging
 import comfy.ldm.lightricks.av_model
-import comfy.context_windows
 from comfy.ldm.modules.diffusionmodules.openaimodel import UNetModel, Timestep
 from comfy.ldm.cascade.stage_c import StageC
 from comfy.ldm.cascade.stage_b import StageB
@@ -1384,6 +1383,7 @@ class WAN21_Vace(WAN21):
 
     def resize_cond_for_context_window(self, cond_key, cond_value, window, x_in, device, retain_index_list=[]):
         if cond_key == "vace_context":
+            import comfy.context_windows
             return comfy.context_windows.slice_cond(cond_value, window, x_in, device, temporal_dim=3, retain_index_list=retain_index_list)
         return super().resize_cond_for_context_window(cond_key, cond_value, window, x_in, device, retain_index_list=retain_index_list)
 
@@ -1441,6 +1441,7 @@ class WAN21_HuMo(WAN21):
 
     def resize_cond_for_context_window(self, cond_key, cond_value, window, x_in, device, retain_index_list=[]):
         if cond_key == "audio_embed":
+            import comfy.context_windows
             return comfy.context_windows.slice_cond(cond_value, window, x_in, device, temporal_dim=1)
         return super().resize_cond_for_context_window(cond_key, cond_value, window, x_in, device, retain_index_list=retain_index_list)
 
@@ -1462,6 +1463,7 @@ class WAN22_Animate(WAN21):
         return out
 
     def resize_cond_for_context_window(self, cond_key, cond_value, window, x_in, device, retain_index_list=[]):
+        import comfy.context_windows
         if cond_key == "face_pixel_values":
             return comfy.context_windows.slice_cond(cond_value, window, x_in, device, temporal_dim=2, temporal_scale=4, temporal_offset=1)
         if cond_key == "pose_latents":
@@ -1506,6 +1508,7 @@ class WAN22_S2V(WAN21):
 
     def resize_cond_for_context_window(self, cond_key, cond_value, window, x_in, device, retain_index_list=[]):
         if cond_key == "audio_embed":
+            import comfy.context_windows
             return comfy.context_windows.slice_cond(cond_value, window, x_in, device, temporal_dim=1)
         return super().resize_cond_for_context_window(cond_key, cond_value, window, x_in, device, retain_index_list=retain_index_list)
 

comfy_api/input/__init__.py

@@ -5,10 +5,6 @@ from comfy_api.latest._input import (
     MaskInput,
     LatentInput,
     VideoInput,
-    CurvePoint,
-    CurveInput,
-    MonotoneCubicCurve,
-    LinearCurve,
 )
 
 __all__ = [
@@ -17,8 +13,4 @@ __all__ = [
     "MaskInput",
     "LatentInput",
     "VideoInput",
-    "CurvePoint",
-    "CurveInput",
-    "MonotoneCubicCurve",
-    "LinearCurve",
 ]

comfy_api/latest/__init__.py

@@ -15,6 +15,7 @@ from comfy_execution.progress import get_progress_state, PreviewImageTuple
 from PIL import Image
 from comfy.cli_args import args
 import numpy as np
+import os
 
 
 class ComfyAPI_latest(ComfyAPIBase):
@@ -25,6 +26,7 @@ class ComfyAPI_latest(ComfyAPIBase):
         super().__init__()
         self.node_replacement = self.NodeReplacement()
         self.execution = self.Execution()
+        self.environment = self.Environment()
         self.caching = self.Caching()
 
     class NodeReplacement(ProxiedSingleton):
@@ -85,6 +87,27 @@ class ComfyAPI_latest(ComfyAPIBase):
                 image=to_display,
             )
 
+    class Environment(ProxiedSingleton):
+        """
+        Query the current execution environment.
+
+        Managed deployments set the ``COMFY_EXECUTION_ENVIRONMENT`` env var
+        so custom nodes can adapt their behaviour at runtime.
+
+        Example::
+
+            from comfy_api.latest import api
+
+            env = api.environment.get()  # "local" | "cloud" | "remote"
+        """
+
+        _VALID = {"local", "cloud", "remote"}
+
+        async def get(self) -> str:
+            """Return the execution environment: ``"local"``, ``"cloud"``, or ``"remote"``."""
+            value = os.environ.get("COMFY_EXECUTION_ENVIRONMENT", "local").lower().strip()
+            return value if value in self._VALID else "local"
+
     class Caching(ProxiedSingleton):
         """
         External cache provider API for sharing cached node outputs

comfy_api/latest/_input/__init__.py

@@ -1,5 +1,4 @@
 from .basic_types import ImageInput, AudioInput, MaskInput, LatentInput
-from .curve_types import CurvePoint, CurveInput, MonotoneCubicCurve, LinearCurve
 from .video_types import VideoInput
 
 __all__ = [
@@ -8,8 +7,4 @@ __all__ = [
     "VideoInput",
     "MaskInput",
     "LatentInput",
-    "CurvePoint",
-    "CurveInput",
-    "MonotoneCubicCurve",
-    "LinearCurve",
 ]

comfy_api/latest/_input/curve_types.py

@@ -1,219 +0,0 @@
-from __future__ import annotations
-
-import logging
-import math
-from abc import ABC, abstractmethod
-import numpy as np
-
-logger = logging.getLogger(__name__)
-
-
-CurvePoint = tuple[float, float]
-
-
-class CurveInput(ABC):
-    """Abstract base class for curve inputs.
-
-    Subclasses represent different curve representations (control-point
-    interpolation, analytical functions, LUT-based, etc.) while exposing a
-    uniform evaluation interface to downstream nodes.
-    """
-
-    @property
-    @abstractmethod
-    def points(self) -> list[CurvePoint]:
-        """The control points that define this curve."""
-
-    @abstractmethod
-    def interp(self, x: float) -> float:
-        """Evaluate the curve at a single *x* value in [0, 1]."""
-
-    def interp_array(self, xs: np.ndarray) -> np.ndarray:
-        """Vectorised evaluation over a numpy array of x values.
-
-        Subclasses should override this for better performance. The default
-        falls back to scalar ``interp`` calls.
-        """
-        return np.fromiter((self.interp(float(x)) for x in xs), dtype=np.float64, count=len(xs))
-
-    def to_lut(self, size: int = 256) -> np.ndarray:
-        """Generate a float64 lookup table of *size* evenly-spaced samples in [0, 1]."""
-        return self.interp_array(np.linspace(0.0, 1.0, size))
-
-    @staticmethod
-    def from_raw(data) -> CurveInput:
-        """Convert raw curve data (dict or point list) to a CurveInput instance.
-
-        Accepts:
-        - A ``CurveInput`` instance (returned as-is).
-        - A dict with ``"points"`` and optional ``"interpolation"`` keys.
-        - A bare list/sequence of ``(x, y)`` pairs (defaults to monotone cubic).
-        """
-        if isinstance(data, CurveInput):
-            return data
-        if isinstance(data, dict):
-            raw_points = data["points"]
-            interpolation = data.get("interpolation", "monotone_cubic")
-        else:
-            raw_points = data
-            interpolation = "monotone_cubic"
-        points = [(float(x), float(y)) for x, y in raw_points]
-        if interpolation == "linear":
-            return LinearCurve(points)
-        if interpolation != "monotone_cubic":
-            logger.warning("Unknown curve interpolation %r, falling back to monotone_cubic", interpolation)
-        return MonotoneCubicCurve(points)
-
-
-class MonotoneCubicCurve(CurveInput):
-    """Monotone cubic Hermite interpolation over control points.
-
-    Mirrors the frontend ``createMonotoneInterpolator`` in
-    ``ComfyUI_frontend/src/components/curve/curveUtils.ts`` so that
-    backend evaluation matches the editor preview exactly.
-
-    All heavy work (sorting, slope computation) happens once at construction.
-    ``interp_array`` is fully vectorised with numpy.
-    """
-
-    def __init__(self, control_points: list[CurvePoint]):
-        sorted_pts = sorted(control_points, key=lambda p: p[0])
-        self._points = [(float(x), float(y)) for x, y in sorted_pts]
-        self._xs = np.array([p[0] for p in self._points], dtype=np.float64)
-        self._ys = np.array([p[1] for p in self._points], dtype=np.float64)
-        self._slopes = self._compute_slopes()
-
-    @property
-    def points(self) -> list[CurvePoint]:
-        return list(self._points)
-
-    def _compute_slopes(self) -> np.ndarray:
-        xs, ys = self._xs, self._ys
-        n = len(xs)
-        if n < 2:
-            return np.zeros(n, dtype=np.float64)
-
-        dx = np.diff(xs)
-        dy = np.diff(ys)
-        dx_safe = np.where(dx == 0, 1.0, dx)
-        deltas = np.where(dx == 0, 0.0, dy / dx_safe)
-
-        slopes = np.empty(n, dtype=np.float64)
-        slopes[0] = deltas[0]
-        slopes[-1] = deltas[-1]
-        for i in range(1, n - 1):
-            if deltas[i - 1] * deltas[i] <= 0:
-                slopes[i] = 0.0
-            else:
-                slopes[i] = (deltas[i - 1] + deltas[i]) / 2
-
-        for i in range(n - 1):
-            if deltas[i] == 0:
-                slopes[i] = 0.0
-                slopes[i + 1] = 0.0
-            else:
-                alpha = slopes[i] / deltas[i]
-                beta = slopes[i + 1] / deltas[i]
-                s = alpha * alpha + beta * beta
-                if s > 9:
-                    t = 3 / math.sqrt(s)
-                    slopes[i] = t * alpha * deltas[i]
-                    slopes[i + 1] = t * beta * deltas[i]
-        return slopes
-
-    def interp(self, x: float) -> float:
-        xs, ys, slopes = self._xs, self._ys, self._slopes
-        n = len(xs)
-        if n == 0:
-            return 0.0
-        if n == 1:
-            return float(ys[0])
-        if x <= xs[0]:
-            return float(ys[0])
-        if x >= xs[-1]:
-            return float(ys[-1])
-
-        hi = int(np.searchsorted(xs, x, side='right'))
-        hi = min(hi, n - 1)
-        lo = hi - 1
-
-        dx = xs[hi] - xs[lo]
-        if dx == 0:
-            return float(ys[lo])
-
-        t = (x - xs[lo]) / dx
-        t2 = t * t
-        t3 = t2 * t
-        h00 = 2 * t3 - 3 * t2 + 1
-        h10 = t3 - 2 * t2 + t
-        h01 = -2 * t3 + 3 * t2
-        h11 = t3 - t2
-        return float(h00 * ys[lo] + h10 * dx * slopes[lo] + h01 * ys[hi] + h11 * dx * slopes[hi])
-
-    def interp_array(self, xs_in: np.ndarray) -> np.ndarray:
-        """Fully vectorised evaluation using numpy."""
-        xs, ys, slopes = self._xs, self._ys, self._slopes
-        n = len(xs)
-        if n == 0:
-            return np.zeros_like(xs_in, dtype=np.float64)
-        if n == 1:
-            return np.full_like(xs_in, ys[0], dtype=np.float64)
-
-        hi = np.searchsorted(xs, xs_in, side='right').clip(1, n - 1)
-        lo = hi - 1
-
-        dx = xs[hi] - xs[lo]
-        dx_safe = np.where(dx == 0, 1.0, dx)
-        t = np.where(dx == 0, 0.0, (xs_in - xs[lo]) / dx_safe)
-        t2 = t * t
-        t3 = t2 * t
-
-        h00 = 2 * t3 - 3 * t2 + 1
-        h10 = t3 - 2 * t2 + t
-        h01 = -2 * t3 + 3 * t2
-        h11 = t3 - t2
-
-        result = h00 * ys[lo] + h10 * dx * slopes[lo] + h01 * ys[hi] + h11 * dx * slopes[hi]
-        result = np.where(xs_in <= xs[0], ys[0], result)
-        result = np.where(xs_in >= xs[-1], ys[-1], result)
-        return result
-
-    def __repr__(self) -> str:
-        return f"MonotoneCubicCurve(points={self._points})"
-
-
-class LinearCurve(CurveInput):
-    """Piecewise linear interpolation over control points.
-
-    Mirrors the frontend ``createLinearInterpolator`` in
-    ``ComfyUI_frontend/src/components/curve/curveUtils.ts``.
-    """
-
-    def __init__(self, control_points: list[CurvePoint]):
-        sorted_pts = sorted(control_points, key=lambda p: p[0])
-        self._points = [(float(x), float(y)) for x, y in sorted_pts]
-        self._xs = np.array([p[0] for p in self._points], dtype=np.float64)
-        self._ys = np.array([p[1] for p in self._points], dtype=np.float64)
-
-    @property
-    def points(self) -> list[CurvePoint]:
-        return list(self._points)
-
-    def interp(self, x: float) -> float:
-        xs, ys = self._xs, self._ys
-        n = len(xs)
-        if n == 0:
-            return 0.0
-        if n == 1:
-            return float(ys[0])
-        return float(np.interp(x, xs, ys))
-
-    def interp_array(self, xs_in: np.ndarray) -> np.ndarray:
-        if len(self._xs) == 0:
-            return np.zeros_like(xs_in, dtype=np.float64)
-        if len(self._xs) == 1:
-            return np.full_like(xs_in, self._ys[0], dtype=np.float64)
-        return np.interp(xs_in, self._xs, self._ys)
-
-    def __repr__(self) -> str:
-        return f"LinearCurve(points={self._points})"

comfy_api/latest/_io.py

@@ -23,7 +23,7 @@ if TYPE_CHECKING:
     from comfy.samplers import CFGGuider, Sampler
     from comfy.sd import CLIP, VAE
     from comfy.sd import StyleModel as StyleModel_
-    from comfy_api.input import VideoInput, CurveInput as CurveInput_
+    from comfy_api.input import VideoInput
 from comfy_api.internal import (_ComfyNodeInternal, _NodeOutputInternal, classproperty, copy_class, first_real_override, is_class,
     prune_dict, shallow_clone_class)
 from comfy_execution.graph_utils import ExecutionBlocker
@@ -1242,9 +1242,8 @@ class BoundingBox(ComfyTypeIO):
 
 @comfytype(io_type="CURVE")
 class Curve(ComfyTypeIO):
-    from comfy_api.input import CurvePoint
-    if TYPE_CHECKING:
-        Type = CurveInput_
+    CurvePoint = tuple[float, float]
+    Type = list[CurvePoint]
 
     class Input(WidgetInput):
         def __init__(self, id: str, display_name: str=None, optional=False, tooltip: str=None,
@@ -1253,18 +1252,6 @@ class Curve(ComfyTypeIO):
             if default is None:
                 self.default = [(0.0, 0.0), (1.0, 1.0)]
 
-        def as_dict(self):
-            d = super().as_dict()
-            if self.default is not None:
-                d["default"] = {"points": [list(p) for p in self.default], "interpolation": "monotone_cubic"}
-            return d
-
-
-@comfytype(io_type="HISTOGRAM")
-class Histogram(ComfyTypeIO):
-    """A histogram represented as a list of bin counts."""
-    Type = list[int]
-
 
 DYNAMIC_INPUT_LOOKUP: dict[str, Callable[[dict[str, Any], dict[str, Any], tuple[str, dict[str, Any]], str, list[str] | None], None]] = {}
 def register_dynamic_input_func(io_type: str, func: Callable[[dict[str, Any], dict[str, Any], tuple[str, dict[str, Any]], str, list[str] | None], None]):
@@ -2253,6 +2240,5 @@ __all__ = [
     "PriceBadge",
     "BoundingBox",
     "Curve",
-    "Histogram",
     "NodeReplace",
 ]

comfy_api_nodes/apis/quiver.py

@@ -1,43 +0,0 @@
-from pydantic import BaseModel, Field
-
-
-class QuiverImageObject(BaseModel):
-    url: str = Field(...)
-
-
-class QuiverTextToSVGRequest(BaseModel):
-    model: str = Field(default="arrow-preview")
-    prompt: str = Field(...)
-    instructions: str | None = Field(default=None)
-    references: list[QuiverImageObject] | None = Field(default=None, max_length=4)
-    temperature: float | None = Field(default=None, ge=0, le=2)
-    top_p: float | None = Field(default=None, ge=0, le=1)
-    presence_penalty: float | None = Field(default=None, ge=-2, le=2)
-
-
-class QuiverImageToSVGRequest(BaseModel):
-    model: str = Field(default="arrow-preview")
-    image: QuiverImageObject = Field(...)
-    auto_crop: bool | None = Field(default=None)
-    target_size: int | None = Field(default=None, ge=128, le=4096)
-    temperature: float | None = Field(default=None, ge=0, le=2)
-    top_p: float | None = Field(default=None, ge=0, le=1)
-    presence_penalty: float | None = Field(default=None, ge=-2, le=2)
-
-
-class QuiverSVGResponseItem(BaseModel):
-    svg: str = Field(...)
-    mime_type: str | None = Field(default="image/svg+xml")
-
-
-class QuiverSVGUsage(BaseModel):
-    total_tokens: int | None = Field(default=None)
-    input_tokens: int | None = Field(default=None)
-    output_tokens: int | None = Field(default=None)
-
-
-class QuiverSVGResponse(BaseModel):
-    id: str | None = Field(default=None)
-    created: int | None = Field(default=None)
-    data: list[QuiverSVGResponseItem] = Field(...)
-    usage: QuiverSVGUsage | None = Field(default=None)

comfy_api_nodes/nodes_quiver.py

@@ -1,291 +0,0 @@
-from io import BytesIO
-
-from typing_extensions import override
-
-from comfy_api.latest import IO, ComfyExtension
-from comfy_api_nodes.apis.quiver import (
-    QuiverImageObject,
-    QuiverImageToSVGRequest,
-    QuiverSVGResponse,
-    QuiverTextToSVGRequest,
-)
-from comfy_api_nodes.util import (
-    ApiEndpoint,
-    sync_op,
-    upload_image_to_comfyapi,
-    validate_string,
-)
-from comfy_extras.nodes_images import SVG
-
-
-class QuiverTextToSVGNode(IO.ComfyNode):
-    @classmethod
-    def define_schema(cls):
-        return IO.Schema(
-            node_id="QuiverTextToSVGNode",
-            display_name="Quiver Text to SVG",
-            category="api node/image/Quiver",
-            description="Generate an SVG from a text prompt using Quiver AI.",
-            inputs=[
-                IO.String.Input(
-                    "prompt",
-                    multiline=True,
-                    default="",
-                    tooltip="Text description of the desired SVG output.",
-                ),
-                IO.String.Input(
-                    "instructions",
-                    multiline=True,
-                    default="",
-                    tooltip="Additional style or formatting guidance.",
-                    optional=True,
-                ),
-                IO.Autogrow.Input(
-                    "reference_images",
-                    template=IO.Autogrow.TemplatePrefix(
-                        IO.Image.Input("image"),
-                        prefix="ref_",
-                        min=0,
-                        max=4,
-                    ),
-                    tooltip="Up to 4 reference images to guide the generation.",
-                    optional=True,
-                ),
-                IO.DynamicCombo.Input(
-                    "model",
-                    options=[
-                        IO.DynamicCombo.Option(
-                            "arrow-preview",
-                            [
-                                IO.Float.Input(
-                                    "temperature",
-                                    default=1.0,
-                                    min=0.0,
-                                    max=2.0,
-                                    step=0.1,
-                                    display_mode=IO.NumberDisplay.slider,
-                                    tooltip="Randomness control. Higher values increase randomness.",
-                                    advanced=True,
-                                ),
-                                IO.Float.Input(
-                                    "top_p",
-                                    default=1.0,
-                                    min=0.05,
-                                    max=1.0,
-                                    step=0.05,
-                                    display_mode=IO.NumberDisplay.slider,
-                                    tooltip="Nucleus sampling parameter.",
-                                    advanced=True,
-                                ),
-                                IO.Float.Input(
-                                    "presence_penalty",
-                                    default=0.0,
-                                    min=-2.0,
-                                    max=2.0,
-                                    step=0.1,
-                                    display_mode=IO.NumberDisplay.slider,
-                                    tooltip="Token presence penalty.",
-                                    advanced=True,
-                                ),
-                            ],
-                        ),
-                    ],
-                    tooltip="Model to use for SVG generation.",
-                ),
-                IO.Int.Input(
-                    "seed",
-                    default=0,
-                    min=0,
-                    max=2147483647,
-                    control_after_generate=True,
-                    tooltip="Seed to determine if node should re-run; "
-                    "actual results are nondeterministic regardless of seed.",
-                ),
-            ],
-            outputs=[
-                IO.SVG.Output(),
-            ],
-            hidden=[
-                IO.Hidden.auth_token_comfy_org,
-                IO.Hidden.api_key_comfy_org,
-                IO.Hidden.unique_id,
-            ],
-            is_api_node=True,
-            price_badge=IO.PriceBadge(
-                expr="""{"type":"usd","usd":0.429}""",
-            ),
-        )
-
-    @classmethod
-    async def execute(
-        cls,
-        prompt: str,
-        model: dict,
-        seed: int,
-        instructions: str = None,
-        reference_images: IO.Autogrow.Type = None,
-    ) -> IO.NodeOutput:
-        validate_string(prompt, strip_whitespace=False, min_length=1)
-
-        references = None
-        if reference_images:
-            references = []
-            for key in reference_images:
-                url = await upload_image_to_comfyapi(cls, reference_images[key])
-                references.append(QuiverImageObject(url=url))
-            if len(references) > 4:
-                raise ValueError("Maximum 4 reference images are allowed.")
-
-        instructions_val = instructions.strip() if instructions else None
-        if instructions_val == "":
-            instructions_val = None
-
-        response = await sync_op(
-            cls,
-            ApiEndpoint(path="/proxy/quiver/v1/svgs/generations", method="POST"),
-            response_model=QuiverSVGResponse,
-            data=QuiverTextToSVGRequest(
-                model=model["model"],
-                prompt=prompt,
-                instructions=instructions_val,
-                references=references,
-                temperature=model.get("temperature"),
-                top_p=model.get("top_p"),
-                presence_penalty=model.get("presence_penalty"),
-            ),
-        )
-
-        svg_data = [BytesIO(item.svg.encode("utf-8")) for item in response.data]
-        return IO.NodeOutput(SVG(svg_data))
-
-
-class QuiverImageToSVGNode(IO.ComfyNode):
-    @classmethod
-    def define_schema(cls):
-        return IO.Schema(
-            node_id="QuiverImageToSVGNode",
-            display_name="Quiver Image to SVG",
-            category="api node/image/Quiver",
-            description="Vectorize a raster image into SVG using Quiver AI.",
-            inputs=[
-                IO.Image.Input(
-                    "image",
-                    tooltip="Input image to vectorize.",
-                ),
-                IO.Boolean.Input(
-                    "auto_crop",
-                    default=False,
-                    tooltip="Automatically crop to the dominant subject.",
-                ),
-                IO.DynamicCombo.Input(
-                    "model",
-                    options=[
-                        IO.DynamicCombo.Option(
-                            "arrow-preview",
-                            [
-                                IO.Int.Input(
-                                    "target_size",
-                                    default=1024,
-                                    min=128,
-                                    max=4096,
-                                    tooltip="Square resize target in pixels.",
-                                ),
-                                IO.Float.Input(
-                                    "temperature",
-                                    default=1.0,
-                                    min=0.0,
-                                    max=2.0,
-                                    step=0.1,
-                                    display_mode=IO.NumberDisplay.slider,
-                                    tooltip="Randomness control. Higher values increase randomness.",
-                                    advanced=True,
-                                ),
-                                IO.Float.Input(
-                                    "top_p",
-                                    default=1.0,
-                                    min=0.05,
-                                    max=1.0,
-                                    step=0.05,
-                                    display_mode=IO.NumberDisplay.slider,
-                                    tooltip="Nucleus sampling parameter.",
-                                    advanced=True,
-                                ),
-                                IO.Float.Input(
-                                    "presence_penalty",
-                                    default=0.0,
-                                    min=-2.0,
-                                    max=2.0,
-                                    step=0.1,
-                                    display_mode=IO.NumberDisplay.slider,
-                                    tooltip="Token presence penalty.",
-                                    advanced=True,
-                                ),
-                            ],
-                        ),
-                    ],
-                    tooltip="Model to use for SVG vectorization.",
-                ),
-                IO.Int.Input(
-                    "seed",
-                    default=0,
-                    min=0,
-                    max=2147483647,
-                    control_after_generate=True,
-                    tooltip="Seed to determine if node should re-run; "
-                    "actual results are nondeterministic regardless of seed.",
-                ),
-            ],
-            outputs=[
-                IO.SVG.Output(),
-            ],
-            hidden=[
-                IO.Hidden.auth_token_comfy_org,
-                IO.Hidden.api_key_comfy_org,
-                IO.Hidden.unique_id,
-            ],
-            is_api_node=True,
-            price_badge=IO.PriceBadge(
-                expr="""{"type":"usd","usd":0.429}""",
-            ),
-        )
-
-    @classmethod
-    async def execute(
-        cls,
-        image,
-        auto_crop: bool,
-        model: dict,
-        seed: int,
-    ) -> IO.NodeOutput:
-        image_url = await upload_image_to_comfyapi(cls, image)
-
-        response = await sync_op(
-            cls,
-            ApiEndpoint(path="/proxy/quiver/v1/svgs/vectorizations", method="POST"),
-            response_model=QuiverSVGResponse,
-            data=QuiverImageToSVGRequest(
-                model=model["model"],
-                image=QuiverImageObject(url=image_url),
-                auto_crop=auto_crop if auto_crop else None,
-                target_size=model.get("target_size"),
-                temperature=model.get("temperature"),
-                top_p=model.get("top_p"),
-                presence_penalty=model.get("presence_penalty"),
-            ),
-        )
-
-        svg_data = [BytesIO(item.svg.encode("utf-8")) for item in response.data]
-        return IO.NodeOutput(SVG(svg_data))
-
-
-class QuiverExtension(ComfyExtension):
-    @override
-    async def get_node_list(self) -> list[type[IO.ComfyNode]]:
-        return [
-            QuiverTextToSVGNode,
-            QuiverImageToSVGNode,
-        ]
-
-
-async def comfy_entrypoint() -> QuiverExtension:
-    return QuiverExtension()

comfy_extras/nodes_canny.py

@@ -3,7 +3,6 @@ from typing_extensions import override
 
 import comfy.model_management
 from comfy_api.latest import ComfyExtension, io
-import torch
 
 
 class Canny(io.ComfyNode):
@@ -30,8 +29,8 @@ class Canny(io.ComfyNode):
 
     @classmethod
     def execute(cls, image, low_threshold, high_threshold) -> io.NodeOutput:
-        output = canny(image.to(device=comfy.model_management.get_torch_device(), dtype=torch.float32).movedim(-1, 1), low_threshold, high_threshold)
-        img_out = output[1].to(device=comfy.model_management.intermediate_device(), dtype=comfy.model_management.intermediate_dtype()).repeat(1, 3, 1, 1).movedim(1, -1)
+        output = canny(image.to(comfy.model_management.get_torch_device()).movedim(-1, 1), low_threshold, high_threshold)
+        img_out = output[1].to(comfy.model_management.intermediate_device()).repeat(1, 3, 1, 1).movedim(1, -1)
         return io.NodeOutput(img_out)
 
 

comfy_extras/nodes_curve.py

@@ -1,42 +0,0 @@
-from __future__ import annotations
-
-from comfy_api.latest import ComfyExtension, io
-from comfy_api.input import CurveInput
-from typing_extensions import override
-
-
-class CurveEditor(io.ComfyNode):
-    @classmethod
-    def define_schema(cls):
-        return io.Schema(
-            node_id="CurveEditor",
-            display_name="Curve Editor",
-            category="utils",
-            inputs=[
-                io.Curve.Input("curve"),
-                io.Histogram.Input("histogram", optional=True),
-            ],
-            outputs=[
-                io.Curve.Output("curve"),
-            ],
-        )
-
-    @classmethod
-    def execute(cls, curve, histogram=None) -> io.NodeOutput:
-        result = CurveInput.from_raw(curve)
-
-        ui = {}
-        if histogram is not None:
-            ui["histogram"] = histogram if isinstance(histogram, list) else list(histogram)
-
-        return io.NodeOutput(result, ui=ui) if ui else io.NodeOutput(result)
-
-
-class CurveExtension(ComfyExtension):
-    @override
-    async def get_node_list(self):
-        return [CurveEditor]
-
-
-async def comfy_entrypoint():
-    return CurveExtension()

comfyui_version.py

@@ -1,3 +1,3 @@
 # This file is automatically generated by the build process when version is
 # updated in pyproject.toml.
-__version__ = "0.18.0"
+__version__ = "0.17.0"

nodes.py

@@ -1966,11 +1966,9 @@ class EmptyImage:
     CATEGORY = "image"
 
     def generate(self, width, height, batch_size=1, color=0):
-        dtype = comfy.model_management.intermediate_dtype()
-        device = comfy.model_management.intermediate_device()
-        r = torch.full([batch_size, height, width, 1], ((color >> 16) & 0xFF) / 0xFF, device=device, dtype=dtype)
-        g = torch.full([batch_size, height, width, 1], ((color >> 8) & 0xFF) / 0xFF, device=device, dtype=dtype)
-        b = torch.full([batch_size, height, width, 1], ((color) & 0xFF) / 0xFF, device=device, dtype=dtype)
+        r = torch.full([batch_size, height, width, 1], ((color >> 16) & 0xFF) / 0xFF)
+        g = torch.full([batch_size, height, width, 1], ((color >> 8) & 0xFF) / 0xFF)
+        b = torch.full([batch_size, height, width, 1], ((color) & 0xFF) / 0xFF)
         return (torch.cat((r, g, b), dim=-1), )
 
 class ImagePadForOutpaint:
@@ -2455,7 +2453,6 @@ async def init_builtin_extra_nodes():
         "nodes_sdpose.py",
         "nodes_math.py",
         "nodes_painter.py",
-        "nodes_curve.py",
     ]
 
     import_failed = []

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ComfyUI"
-version = "0.18.0"
+version = "0.17.0"
 readme = "README.md"
 license = { file = "LICENSE" }
 requires-python = ">=3.10"