...

backend/src/nodes/impl/stable_diffusion/types.py

@@ -1,9 +1,9 @@
-from comfy import StableDiffusionModel, VAEModel, CLIPModel, LatentImage, Conditioning
+from comfy import CLIPModel, Conditioning, LatentImage, StableDiffusionModel, VAEModel
 
 __all__ = [
     "StableDiffusionModel",
     "VAEModel",
     "CLIPModel",
     "LatentImage",
     "Conditioning",
-]
+]

backend/src/nodes/nodes/builtin_categories.py

@@ -6,10 +6,10 @@
 from .image_filter import category as ImageFilterCategory
 from .image_utility import category as ImageUtilityCategory
 from .material_textures import category as MaterialTexturesCategory
-from .stable_diffusion import category as StableDiffusionCategory
 from .ncnn import category as NCNNCategory
 from .onnx import category as ONNXCategory
 from .pytorch import category as PyTorchCategory
+from .stable_diffusion import category as StableDiffusionCategory
 from .utility import category as UtilityCategory
 
 builtin_categories = [

backend/src/nodes/nodes/stable_diffusion/clip_encode.py

@@ -1,24 +1,25 @@
 from __future__ import annotations
 
-from typing import Tuple, Optional
+from typing import Optional
 
 from ...impl.stable_diffusion.types import CLIPModel, Conditioning
-
 from ...node_base import NodeBase
 from ...node_factory import NodeFactory
 from ...properties.inputs import TextAreaInput
 from ...properties.inputs.stable_diffusion_inputs import CLIPModelInput
-from . import category as StableDiffusionCategory
 from ...properties.outputs.stable_diffusion_outputs import ConditioningOutput
+from . import category as StableDiffusionCategory
 
 
 @NodeFactory.register("chainner:stable_diffusion:clip_encode")
 class CLIPEncodeNode(NodeBase):
     def __init__(self):
         super().__init__()
         self.description = ""
-        self.inputs = [CLIPModelInput(),
-            TextAreaInput("Prompt").make_optional(),]
+        self.inputs = [
+            CLIPModelInput(),
+            TextAreaInput("Prompt").make_optional(),
+        ]
         self.outputs = [
             ConditioningOutput(),
         ]
@@ -31,4 +32,4 @@ def __init__(self):
     def run(self, clip: CLIPModel, prompt: Optional[str]) -> Conditioning:
         prompt = prompt or ""
         out = clip.encode(prompt)
-        return out
+        return out

backend/src/nodes/nodes/stable_diffusion/conditioning_compose.py

@@ -0,0 +1,30 @@
+from __future__ import annotations
+
+from ...impl.stable_diffusion.types import Conditioning
+from ...node_base import NodeBase
+from ...node_factory import NodeFactory
+from ...properties.inputs.stable_diffusion_inputs import ConditioningInput
+from ...properties.outputs.stable_diffusion_outputs import ConditioningOutput
+from . import category as StableDiffusionCategory
+
+
+@NodeFactory.register("chainner:stable_diffusion:conditioning_compose")
+class ConditinoingComposeNode(NodeBase):
+    def __init__(self):
+        super().__init__()
+        self.description = ""
+        self.inputs = [
+            ConditioningInput(),
+            ConditioningInput(),
+        ]
+        self.outputs = [
+            ConditioningOutput(),
+        ]
+
+        self.category = StableDiffusionCategory
+        self.name = "Compose"
+        self.icon = "PyTorch"
+        self.sub = "Conditioning"
+
+    def run(self, a: Conditioning, b: Conditioning) -> Conditioning:
+        return Conditioning.combine([a, b])

backend/src/nodes/nodes/stable_diffusion/conditioning_set_area.py

@@ -0,0 +1,82 @@
+from __future__ import annotations
+
+from ...impl.stable_diffusion.types import Conditioning
+from ...node_base import NodeBase
+from ...node_factory import NodeFactory
+from ...properties.inputs import SliderInput
+from ...properties.inputs.stable_diffusion_inputs import ConditioningInput
+from ...properties.outputs.stable_diffusion_outputs import ConditioningOutput
+from . import category as StableDiffusionCategory
+
+
+@NodeFactory.register("chainner:stable_diffusion:conditioning_set_area")
+class ConditinoingComposeNode(NodeBase):
+    def __init__(self):
+        super().__init__()
+        self.description = ""
+        self.inputs = [
+            ConditioningInput(),
+            SliderInput(
+                "strength",
+                minimum=0,
+                maximum=10,
+                default=1,
+                slider_step=0.01,
+                controls_step=0.01,
+            ),
+            SliderInput(
+                "width",
+                unit="px",
+                minimum=64,
+                maximum=4096,
+                default=512,
+                slider_step=64,
+                controls_step=64,
+            ),
+            SliderInput(
+                "height",
+                unit="px",
+                minimum=64,
+                maximum=4096,
+                default=512,
+                slider_step=64,
+                controls_step=64,
+            ),
+            SliderInput(
+                "x",
+                unit="px",
+                minimum=64,
+                maximum=4096,
+                default=512,
+                slider_step=64,
+                controls_step=64,
+            ),
+            SliderInput(
+                "y",
+                unit="px",
+                minimum=64,
+                maximum=4096,
+                default=512,
+                slider_step=64,
+                controls_step=64,
+            ),
+        ]
+        self.outputs = [
+            ConditioningOutput(),
+        ]
+
+        self.category = StableDiffusionCategory
+        self.name = "Set Area"
+        self.icon = "PyTorch"
+        self.sub = "Conditioning"
+
+    def run(
+        self,
+        cond: Conditioning,
+        strength: float,
+        width: int,
+        height: int,
+        x: int,
+        y: int,
+    ) -> Conditioning:
+        return cond.set_area(width, height, x, y, strength)

backend/src/nodes/nodes/stable_diffusion/empty_latent_image.py

@@ -1,16 +1,11 @@
 from __future__ import annotations
 
-from typing import Tuple, Optional
-
-
-from ...impl.stable_diffusion.types import CLIPModel, Conditioning, LatentImage
-
+from ...impl.stable_diffusion.types import LatentImage
 from ...node_base import NodeBase
 from ...node_factory import NodeFactory
-from ...properties.inputs import TextAreaInput, SliderInput
-from ...properties.inputs.stable_diffusion_inputs import CLIPModelInput
+from ...properties.inputs import SliderInput
+from ...properties.outputs.stable_diffusion_outputs import LatentImageOutput
 from . import category as StableDiffusionCategory
-from ...properties.outputs.stable_diffusion_outputs import ConditioningOutput, LatentImageOutput
 
 
 @NodeFactory.register("chainner:stable_diffusion:empty_latent_image")
@@ -47,4 +42,4 @@ def __init__(self):
 
     def run(self, width: int, height: int) -> LatentImage:
         img = LatentImage.empty(width, height)
-        return img
+        return img

backend/src/nodes/nodes/stable_diffusion/k_sampler.py

@@ -1,18 +1,22 @@
 from __future__ import annotations
 
-from typing import Tuple, Optional
-
 import comfy
 
-from ...impl.stable_diffusion.types import StableDiffusionModel, Conditioning, LatentImage
-
+from ...impl.stable_diffusion.types import (
+    Conditioning,
+    LatentImage,
+    StableDiffusionModel,
+)
 from ...node_base import NodeBase, group
 from ...node_factory import NodeFactory
-from ...properties.inputs import TextAreaInput, SliderInput, NumberInput, EnumInput
-from ...properties.inputs.stable_diffusion_inputs import CLIPModelInput, LatentImageInput, ConditioningInput, \
-    StableDiffusionModelInput
+from ...properties.inputs import EnumInput, NumberInput, SliderInput
+from ...properties.inputs.stable_diffusion_inputs import (
+    ConditioningInput,
+    LatentImageInput,
+    StableDiffusionModelInput,
+)
+from ...properties.outputs.stable_diffusion_outputs import LatentImageOutput
 from . import category as StableDiffusionCategory
-from ...properties.outputs.stable_diffusion_outputs import ConditioningOutput, LatentImageOutput
 
 
 @NodeFactory.register("chainner:stable_diffusion:k_sampler")
@@ -28,7 +32,7 @@ def __init__(self):
             SliderInput(
                 "Denoising Strength",
                 minimum=0,
-                default=0.75,
+                default=1,
                 maximum=1,
                 slider_step=0.01,
                 controls_step=0.1,
@@ -60,18 +64,33 @@ def __init__(self):
         ]
 
         self.category = StableDiffusionCategory
-        self.name = "K-Sampler"
+        self.name = "Sample"
         self.icon = "PyTorch"
         self.sub = "Latent"
 
-    def run(self, model: StableDiffusionModel, positive: Conditioning, negative: Conditioning, latent_image: LatentImage,
-            denoising_strength: float, seed: int, steps: int, sampler: comfy.Sampler, scheduler: comfy.Scheduler,
-            cfg_scale: float) -> LatentImage:
-
+    def run(
+        self,
+        model: StableDiffusionModel,
+        positive: Conditioning,
+        negative: Conditioning,
+        latent_image: LatentImage,
+        denoising_strength: float,
+        seed: int,
+        steps: int,
+        sampler: comfy.Sampler,
+        scheduler: comfy.Scheduler,
+        cfg_scale: float,
+    ) -> LatentImage:
         img = model.sample(
-            positive=positive, negative=negative, latent_image=latent_image,
-            seed=seed, steps=steps, cfg_scale=cfg_scale,
-            sampler=sampler, scheduler=scheduler, denoise_strength=denoising_strength,
+            positive=positive,
+            negative=negative,
+            latent_image=latent_image,
+            seed=seed,
+            steps=steps,
+            cfg_scale=cfg_scale,
+            sampler=sampler,
+            scheduler=scheduler,
+            denoise_strength=denoising_strength,
         )
 
         return img