Merge branch 'master' into beta

README.md

@@ -11,7 +11,7 @@ This ui will let you design and execute advanced stable diffusion pipelines usin
 
 ## Features
 - Nodes/graph/flowchart interface to experiment and create complex Stable Diffusion workflows without needing to code anything.
-- Fully supports SD1.x, SD2.x and SDXL
+- Fully supports SD1.x, SD2.x, [SDXL](https://comfyanonymous.github.io/ComfyUI_examples/sdxl/) and [Stable Video Diffusion](https://comfyanonymous.github.io/ComfyUI_examples/video/)
 - Asynchronous Queue system
 - Many optimizations: Only re-executes the parts of the workflow that changes between executions.
 - Command line option: ```--lowvram``` to make it work on GPUs with less than 3GB vram (enabled automatically on GPUs with low vram)

comfy_extras/nodes_video_model.py

@@ -21,7 +21,7 @@ def load_checkpoint(self, ckpt_name, output_vae=True, output_clip=True):
         return (out[0], out[3], out[2])
 
 
-class SDV_img2vid_Conditioning:
+class SVD_img2vid_Conditioning:
     @classmethod
     def INPUT_TYPES(s):
         return {"required": { "clip_vision": ("CLIP_VISION",),
@@ -80,7 +80,7 @@ def linear_cfg(args):
 
 NODE_CLASS_MAPPINGS = {
     "ImageOnlyCheckpointLoader": ImageOnlyCheckpointLoader,
-    "SDV_img2vid_Conditioning": SDV_img2vid_Conditioning,
+    "SVD_img2vid_Conditioning": SVD_img2vid_Conditioning,
     "VideoLinearCFGGuidance": VideoLinearCFGGuidance,
 }
 

web/scripts/app.js

@@ -1532,6 +1532,7 @@ export class ComfyApp {
 			// Patch T2IAdapterLoader to ControlNetLoader since they are the same node now
 			if (n.type == "T2IAdapterLoader") n.type = "ControlNetLoader";
 			if (n.type == "ConditioningAverage ") n.type = "ConditioningAverage"; //typo fix
+			if (n.type == "SDV_img2vid_Conditioning") n.type = "SVD_img2vid_Conditioning"; //typo fix
 
 			// Find missing node types
 			if (!(n.type in LiteGraph.registered_node_types)) {