# ComfyUI: English Built-in Nodes Nodes Model ## Model ### Conditioning - [AudioEncoderEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/AudioEncoderEncode.md): The AudioEncoderEncode node converts audio data into an encoded representation using an audio encoder model. - [CLIPSetLastLayer - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipSetLastLayer.md): CLIP Set Last Layer is a core node in ComfyUI for controlling the processing depth of CLIP models. - [CLIPTextEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipTextEncode.md): CLIP Text Encode (CLIPTextEncode) acts as a translator, converting your text descriptions into a format that AI can understand. - [CLIPTextEncodeControlnet - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPTextEncodeControlnet.md): The CLIPTextEncodeControlnet node processes a text prompt using a CLIP model and combines the resulting text encoding with existing conditioning data. - [CLIPVisionEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipVisionEncode.md): The CLIP Vision Encode node is an image encoding node in ComfyUI, used to convert input images into visual feature vectors through the CLIP Vision model. - [InpaintModelConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/InpaintModelConditioning.md): The InpaintModelConditioning node is designed to facilitate the conditioning process for inpainting models, enabling the integration and manipulation of various conditioning inputs - [NormalizeVideoLatentStart - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/NormalizeVideoLatentStart.md): This node adjusts the first few frames of a video latent to make them look more like the frames that come after. - [PiDConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/PiDConditioning.md): Attaches a latent image and a degrade sigma value to a CONDITIONING data. - [ReferenceLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ReferenceLatent.md): This node sets the guiding latent for an edit model. - [ReferenceTimbreAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ReferenceTimbreAudio.md): This node sets a reference audio timbre for use in the "ace step 1.5" process. - [SeedVR2Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SeedVR2Conditioning.md): This node builds positive and negative conditioning from a VAE latent for use with the SeedVR2 model. - [StyleModelApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StyleModelApply.md): This node applies a style model to a given conditioning, enhancing or altering its style based on the output of a CLIP vision model. - [T5TokenizerOptions - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/T5TokenizerOptions.md): The T5TokenizerOptions node configures tokenizer settings for various T5 model types. - [unCLIPConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/unCLIPConditioning.md): This node is designed to integrate CLIP vision outputs into the conditioning process, adjusting the influence of these outputs based on specified strength and noise augmentation pa #### Ace - [TextEncodeAceStepAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeAceStepAudio.md): The TextEncodeAceStepAudio node processes text inputs for audio conditioning by combining tags and lyrics into tokens, then encoding them with adjustable lyrics strength. - [TextEncodeAceStepAudio1.5 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeAceStepAudio1.5.md): The TextEncodeAceStepAudio1.5 node prepares text and audio-related metadata for use with the AceStepAudio 1.5 model. #### Autoregressive - [ARVideoI2V - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ARVideoI2V.md): Documentation for ARVideoI2V node. #### Bernini - [BerniniConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/BerniniConditioning.md): The BerniniConditioning node prepares video and image conditioning data for the Wan2.2-A14B model. #### Boogu - [TextEncodeBooguEdit - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeBooguEdit.md): This node prepares conditioning for image editing with Boogu. #### Controlnet - [ControlNetApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ControlNetApply.md): Using controlNet requires preprocessing of input images. - [ControlNetApplyAdvanced - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ControlNetApplyAdvanced.md): This node applies advanced control net transformations to conditioning data based on an image and a control net model. - [ControlNetApplySD3 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ControlNetApplySD3.md): This node applies ControlNet guidance to Stable Diffusion 3 conditioning. - [ControlNetInpaintingAliMamaApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ControlNetInpaintingAliMamaApply.md): The ControlNetInpaintingAliMamaApply node applies ControlNet conditioning for inpainting tasks by combining positive and negative conditioning with a control image and mask. - [SetUnionControlNetType - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SetUnionControlNetType.md): The SetUnionControlNetType node lets you choose which control type a control network uses. #### Cosmos - [CosmosImageToVideoLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CosmosImageToVideoLatent.md): The CosmosImageToVideoLatent node creates a video latent representation from input images. - [CosmosPredict2ImageToVideoLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CosmosPredict2ImageToVideoLatent.md): CosmosPredict2ImageToVideoLatent creates video latent representations from images for video generation. #### Flux - [CLIPTextEncodeFlux - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipTextEncodeFlux.md): CLIPTextEncodeFlux is an advanced text encoding node designed for the Flux architecture. - [FluxDisableGuidance - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FluxDisableGuidance.md): This node completely disables the guidance embed on Flux and Flux-like models. - [FluxGuidance - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FluxGuidance.md): Documentation for FluxGuidance node. - [FluxKontextImageScale - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FluxKontextImageScale.md): This node scales the input image to an optimal size used during Flux Kontext model training using the Lanczos algorithm, based on the input image's aspect ratio. - [FluxKontextMultiReferenceLatentMethod - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FluxKontextMultiReferenceLatentMethod.md): The FluxKontextMultiReferenceLatentMethod node modifies conditioning data by setting a specific reference latents method. #### Gligen - [GLIGENTextBoxApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/GLIGENTextBoxApply.md): The GLIGENTextBoxApply node is designed to integrate text-based conditioning into a generative model's input, specifically by applying text box parameters and encoding them using a #### Hidream - [CLIPTextEncodeHiDream - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPTextEncodeHiDream.md): The CLIPTextEncodeHiDream node processes four separate text inputs using different language models (CLIP-L, CLIP-G, T5-XXL, and LLaMA) and combines them into a single conditioning - [HiDreamO1ReferenceImages - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HiDreamO1ReferenceImages.md): This node attaches reference images to both positive and negative conditioning, so downstream nodes can use them to guide generation. #### Hunyuan 3d - [Hunyuan3Dv2Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Hunyuan3Dv2Conditioning.md): The Hunyuan3Dv2Conditioning node processes CLIP vision output to generate conditioning data for 3D models. - [Hunyuan3Dv2ConditioningMultiView - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Hunyuan3Dv2ConditioningMultiView.md): The Hunyuan3Dv2ConditioningMultiView node combines CLIP vision outputs from up to four views (front, left, back, and right) into a single multi-view conditioning. #### Hunyuan Image - [CLIPTextEncodeHunyuanDiT - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipTextEncodeHunyuanDit.md): The CLIPTextEncodeHunyuanDiT node converts text descriptions into a format that the HunyuanDiT model can understand. #### Hunyuan Video - [HunyuanImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HunyuanImageToVideo.md): The HunyuanImageToVideo node converts images into video latent representations using the Hunyuan video model. - [HunyuanRefinerLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HunyuanRefinerLatent.md): The HunyuanRefinerLatent node prepares conditioning and latent data for the Hunyuan video refinement process. - [HunyuanVideo15ImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HunyuanVideo15ImageToVideo.md): The HunyuanVideo15ImageToVideo node prepares conditioning and latent space data for video generation based on the HunyuanVideo 1.5 model. - [HunyuanVideo15SuperResolution - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HunyuanVideo15SuperResolution.md): The HunyuanVideo15SuperResolution node prepares conditioning data for a video super-resolution process. - [TextEncodeHunyuanVideo_ImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeHunyuanVideo_ImageToVideo.md): The TextEncodeHunyuanVideoImageToVideo node creates conditioning data for image-to-video generation by combining a text prompt with visual information from a reference image. #### Instructpix2pix - [InstructPixToPixConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/InstructPixToPixConditioning.md): The InstructPixToPixConditioning node prepares conditioning data for InstructPix2Pix image editing by combining positive and negative text prompts with image data. #### Joyimage - [TextEncodeJoyImageEdit - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeJoyImageEdit.md): This node encodes a text prompt and optional images into conditioning data for use with JoyImage models. #### Kandinsky - [CLIPTextEncodeKandinsky5 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPTextEncodeKandinsky5.md): The CLIP Text Encode (Kandinsky 5) node prepares text prompts for use with the Kandinsky 5 model. - [Kandinsky5ImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Kandinsky5ImageToVideo.md): The Kandinsky5ImageToVideo node prepares conditioning and latent space data for video generation using the Kandinsky model. #### Lotus - [LotusConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LotusConditioning.md): The LotusConditioning node provides pre-computed conditioning embeddings for the Lotus model. #### Ltxv - [GetICLoRAParameters - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/GetICLoRAParameters.md): This node reads the metadata from a LoRA-loaded model to extract IC-LoRA parameters, such as the reference downscale factor. - [LTXVAddGuide - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVAddGuide.md): The LTXVAddGuide node adds video conditioning guidance to latent sequences by encoding input images or videos and incorporating them as keyframes into the conditioning data. - [LTXVConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVConditioning.md): The LTXVConditioning node adds frame rate information to both positive and negative conditioning inputs for video generation models. - [LTXVCropGuides - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVCropGuides.md): The LTXVCropGuides node processes conditioning and latent inputs for video generation by removing keyframe information and adjusting the latent dimensions. - [LTXVDurationPredictor - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVDurationPredictor.md): This node predicts the natural shot duration for a text prompt using an LTX 2.4 duration head loaded with ModelPatchLoader, then snaps the result to the VAE's 8k+1 frame grid. - [LTXVImgToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVImgToVideo.md): LTXVImgToVideo converts an input image into a video latent representation for video generation models. - [LTXVImgToVideoInplace - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVImgToVideoInplace.md): LTXVImgToVideoInplace encodes an input image into the latent space and places those encoded frames at the start of an existing latent video. - [LTXVReferenceAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVReferenceAudio.md): LTXV Reference Audio transfers the voice identity of a speaker from a reference audio clip to generated audio. #### Lumina - [CLIPTextEncodeLumina2 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPTextEncodeLumina2.md): This node encodes a system prompt and a user prompt using a CLIP model into an embedding that can be used to guide the diffusion model towards generating specific images. #### Mage - [TextEncodeMageFlowEdit - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeMageFlowEdit.md): Documentation for TextEncodeMageFlowEdit node. #### Minimax - [MiniMaxH3AddGuide - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MiniMaxH3AddGuide.md): This node anchors an image, a short clip, audio, or a clip with its soundtrack at any chosen frame of a MiniMax H3 video. - [MiniMaxH3ImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MiniMaxH3ImageToVideo.md): This node prepares the conditioning and empty latent needed to generate a video with the MiniMax H3 model. - [MiniMaxH3ReferenceToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MiniMaxH3ReferenceToVideo.md): MiniMax H3 Reference to Video creates the text conditioning and the empty audio-video latent needed for MiniMax H3 reference-to-video generation. #### Minimax Music - [MiniMaxMusic3TextEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MiniMaxMusic3TextEncode.md): MiniMax Music3 Text Encode uses a MiniMax Music3 CLIP model to convert text captions and lyrics into an acoustic conditioning sequence for music generation. #### Photomaker - [PhotoMakerEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/PhotoMakerEncode.md): The PhotoMakerEncode node combines a reference image with a text prompt to create conditioning data for image generation. #### Pixart - [CLIPTextEncodePixArtAlpha - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPTextEncodePixArtAlpha.md): Encodes text and sets the resolution conditioning for PixArt Alpha. #### Qwen Image - [TextEncodeQwenImageEdit - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeQwenImageEdit.md): The TextEncodeQwenImageEdit node converts text prompts and optional images into conditioning data for image generation or editing. - [TextEncodeQwenImageEditPlus - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeQwenImageEditPlus.md): The TextEncodeQwenImageEditPlus node processes text prompts and optional images to generate conditioning data for image generation or editing tasks. #### Stable Audio - [ConditioningStableAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningStableAudio.md): The ConditioningStableAudio node adds timing information to both positive and negative conditioning inputs for audio generation. #### Stable Cascade - [StableCascade_StageB_Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StableCascade_StageB_Conditioning.md): The StableCascadeStageBConditioning node prepares conditioning data for Stable Cascade Stage B generation by combining existing conditioning information with prior latent represent #### Stable Diffusion - [CLIPTextEncodeSD3 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPTextEncodeSD3.md): CLIPTextEncodeSD3 processes text inputs for Stable Diffusion 3 models by encoding multiple text prompts using different CLIP models. - [CLIPTextEncodeSDXL - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipTextEncodeSdxl.md): This node is designed to encode text input using a CLIP model specifically customized for the SDXL architecture. - [CLIPTextEncodeSDXLRefiner - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipTextEncodeSdxlRefiner.md): This node is specifically designed for the SDXL Refiner model to convert text prompts into conditioning information by incorporating aesthetic scores and dimensional information to #### Stable Diffusion Upscaler - [SD_4XUpscale_Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SD_4XUpscale_Conditioning.md): The SD4XUpscaleConditioning node prepares conditioning data for upscaling images using diffusion models. #### Stable Video 3d - [SV3D_Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SV3D_Conditioning.md): SV3DConditioning prepares conditioning data for 3D video generation using the SV3D model. #### Stable Zero123 - [StableZero123_Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StableZero123_Conditioning.md): The StableZero123Conditioning node processes an input image and camera angles to generate conditioning data and latent representations for 3D model generation. - [StableZero123_Conditioning_Batched - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StableZero123_Conditioning_Batched.md): The StableZero123ConditioningBatched node prepares conditioning data for generating a 3D model from a single input image. #### Transform - [ConditioningConcat - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningConcat.md): The ConditioningConcat node is designed to concatenate conditioning vectors, specifically merging the 'conditioningfrom' vector into the 'conditioningto' vector. - [ConditioningMultiply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningMultiply.md): This node multiplies the conditioning values by a specified factor, allowing you to scale the influence of the conditioning on the generation process. - [ConditioningSetArea - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningSetArea.md): This node is designed to modify the conditioning information by setting specific areas within the conditioning context. - [ConditioningSetAreaPercentage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningSetAreaPercentage.md): The ConditioningSetAreaPercentage node specializes in adjusting the area of influence for conditioning elements based on percentage values. - [ConditioningSetAreaPercentageVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningSetAreaPercentageVideo.md): The ConditioningSetAreaPercentageVideo node modifies conditioning data by defining a specific area and temporal region for video generation. - [ConditioningSetAreaStrength - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningSetAreaStrength.md): This node is designed to modify the strength attribute of a given conditioning set, allowing for the adjustment of the influence or intensity of the conditioning on the generation - [ConditioningSetMask - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningSetMask.md): This node is designed to modify the conditioning of a generative model by applying a mask with a specified strength to certain areas. - [ConditioningSetTimestepRange - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningSetTimestepRange.md): This node is designed to adjust the temporal aspect of conditioning by setting a specific range of timesteps. - [ConditioningZeroOut - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ConditioningZeroOut.md): This node zeroes out specific elements within the conditioning data structure, effectively neutralizing their influence in subsequent processing steps. #### Trellis2 - [Pixal3DConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Pixal3DConditioning.md): This node prepares image conditioning for the Trellis2 3D generation pipeline. - [Trellis2Conditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Trellis2Conditioning.md): Trellis2Conditioning converts an input image into conditioning data for the TRELLIS.2 model. - [Trellis2ShapeStage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Trellis2ShapeStage.md): This node sets up the first shape-generation sampling pass of the Trellis2 pipeline. - [Trellis2TextureStage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Trellis2TextureStage.md): This node sets up the texture-stage sampling pass for Trellis2 generation. - [Trellis2UpsampleStage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Trellis2UpsampleStage.md): This node takes the 512-resolution shape latent produced by the first shape-stage sampling pass, upscales it to a higher target resolution, and prepares the conditioning and latent #### Triposplat - [TripoSplatConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TripoSplatConditioning.md): This node encodes an input image using the DINOv3 image encoder and the Flux2 VAE to create positive and negative conditioning data for the TripoSplat model. - [TripoSplatPreprocessImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TripoSplatPreprocessImage.md): This node crops each input image to a centered square on a black background and adds padding to reach the specified output size. #### Void - [VOIDInpaintConditioning - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VOIDInpaintConditioning.md): The VOIDInpaintConditioning node prepares the conditioning data needed for inpainting with CogVideoX models. #### Wan - [Wan22ImageToVideoLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Wan22ImageToVideoLatent.md): Wan22ImageToVideoLatent creates video latent representations from images. - [WanFirstLastFrameToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanFirstLastFrameToVideo.md): The WanFirstLastFrameToVideo node prepares conditioning for video generation by combining a start frame and an end frame with text prompts. - [WanImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanImageToVideo.md): The WanImageToVideo node prepares conditioning and latent representations for video generation tasks. ##### Animate - [WanAnimate2Cache - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanAnimate2Cache.md): Caches the pose-video's per-block activations once so they do not need to be recomputed on every sampling step, which roughly halves generation time. - [WanAnimate2ToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanAnimate2ToVideo.md): WanAnimate2ToVideo animates a character from a reference image by transferring the facial expressions, body motion, and hand gestures from a separate pose video. - [WanAnimateToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanAnimateToVideo.md): WanAnimateToVideo prepares conditioning data and an initial latent for generating animated videos with Wan, using inputs such as a reference image, pose, face, background, and opti ##### Camera - [WanCameraEmbedding - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanCameraEmbedding.md): The WanCameraEmbedding node generates camera trajectory embeddings using Plücker embeddings based on camera motion parameters. - [WanCameraImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanCameraImageToVideo.md): The WanCameraImageToVideo node prepares conditioning and latent data for video generation from images. ##### Dancer - [WanDancerEncodeAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanDancerEncodeAudio.md): This node processes an audio input to extract features that can be used to guide a video generation model. - [WanDancerVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanDancerVideo.md): The WanDancerVideo node prepares conditioning data and an empty latent tensor for video generation with the WanDancer model. ##### Fun Control - [Wan22FunControlToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Wan22FunControlToVideo.md): The Wan22FunControlToVideo node prepares conditioning data and an empty latent tensor for video generation with the Wan video model. - [WanFunControlToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanFunControlToVideo.md): This node was added to support the Alibaba Wan Fun Control model for video generation, and was added after [this commit](https://github.com/comfyanonymous/ComfyUI/commit/3661c833bc ##### Fun Inpaint - [WanFunInpaintToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanFunInpaintToVideo.md): The WanFunInpaintToVideo node creates video sequences by inpainting between start and end images. ##### Humo - [WanHuMoImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanHuMoImageToVideo.md): The WanHuMoImageToVideo node converts images to video sequences by generating latent representations for video frames. ##### Infinite Talk - [WanInfiniteTalkToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanInfiniteTalkToVideo.md): WanInfiniteTalkToVideo generates video sequences from audio input. ##### Move - [GenerateTracks - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/GenerateTracks.md): The GenerateTracks node creates multiple parallel motion paths (tracks) for video generation. - [WanMoveConcatTrack - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanMoveConcatTrack.md): The WanMoveConcatTrack node combines two sets of motion tracking data into a single, longer sequence. - [WanMoveTracksFromCoords - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanMoveTracksFromCoords.md): The WanMoveTracksFromCoords node creates motion tracks from a JSON-formatted string of coordinates. - [WanMoveTrackToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanMoveTrackToVideo.md): The WanMoveTrackToVideo node prepares conditioning and latent space data for video generation, incorporating optional motion tracking information. - [WanMoveVisualizeTracks - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanMoveVisualizeTracks.md): The WanMoveVisualizeTracks node overlays motion tracking data onto a sequence of images or video frames. - [WanTrackToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanTrackToVideo.md): The WanTrackToVideo node converts motion tracking data into video sequences by processing track points and generating corresponding video frames. ##### Phantom Subject - [WanPhantomSubjectToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanPhantomSubjectToVideo.md): The WanPhantomSubjectToVideo node prepares conditioning data and a latent for Wan video generation. ##### Scail - [SCAIL2ColoredMask - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SCAIL2ColoredMask.md): This node renders SAM3 tracking data into colored masks that are consumed by the WanSCAILToVideo node. - [WanSCAILToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanSCAILToVideo.md): The WanSCAILToVideo node prepares conditioning and an empty latent space for video generation with SCAIL and SCAIL-2 video models. ##### Sound - [WanSoundImageToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanSoundImageToVideo.md): The WanSoundImageToVideo node generates video content from images with optional audio conditioning. - [WanSoundImageToVideoExtend - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanSoundImageToVideoExtend.md): The WanSoundImageToVideoExtend node extends an existing video latent by generating additional frames, optionally guided by audio, a reference image, and a control video. ##### Vace - [WanVaceToVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanVaceToVideo.md): The WanVaceToVideo node prepares video conditioning data for video generation models. #### Z Image - [TextEncodeZImageOmni - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TextEncodeZImageOmni.md): TextEncodeZImageOmni encodes a text prompt together with up to three optional reference images into a conditioning format for image generation models. ### Latent - [EmptyLatentAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyLatentAudio.md): Empty Latent Audio creates an empty latent tensor for audio processing. - [EmptyLatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyLatentImage.md): The EmptyLatentImage node is designed to generate a blank latent space representation with specified dimensions and batch size. - [LatentComposite - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentComposite.md): The LatentComposite node is designed to blend or merge two latent representations into a single output. - [LatentCompositeMasked - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentCompositeMasked.md): The LatentCompositeMasked node is designed for blending two latent representations together at specified coordinates, optionally using a mask for more controlled compositing. - [LatentUpscale - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentUpscale.md): The LatentUpscale node is designed for upscaling latent representations of images. - [LatentUpscaleBy - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentUpscaleBy.md): The LatentUpscaleBy node is designed for upscaling latent representations of images. - [LoadLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoadLatent.md): The LoadLatent node loads previously saved latent representations from .latent files in the input directory. - [SaveLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SaveLatent.md): SaveLatent saves latent tensors to disk as .latent files so they can be reused or shared later. - [SetLatentNoiseMask - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SetLatentNoiseMask.md): This node is designed to apply a noise mask to a set of latent samples. - [TrimVideoLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TrimVideoLatent.md): The TrimVideoLatent node removes frames from the beginning of a video latent representation. - [VAEDecode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEDecode.md): The VAEDecode node is designed for decoding latent representations into images using a specified Variational Autoencoder (VAE). - [VAEDecodeAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEDecodeAudio.md): The VAEDecodeAudio node converts latent representations back into audio waveforms using a Variational Autoencoder. - [VAEDecodeAudioTiled - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEDecodeAudioTiled.md): This node converts a compressed audio representation (latent samples) back into an audio waveform using a Variational Autoencoder (VAE). - [VAEDecodeTiled - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEDecodeTiled.md): The VAEDecodeTiled node decodes latent representations into images using a tiled approach to handle large images efficiently. - [VAEEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEEncode.md): This node is designed for encoding images into a latent space representation using a specified VAE model. - [VAEEncodeAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEEncodeAudio.md): The VAEEncodeAudio node converts audio data into a latent representation using a Variational Autoencoder (VAE). - [VAEEncodeForInpaint - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEEncodeForInpaint.md): This node is designed for encoding images into a latent representation suitable for inpainting tasks, incorporating additional preprocessing steps to adjust the input image and mas - [VAEEncodeTiled - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEEncodeTiled.md): VAEEncodeTiled processes images by breaking them into smaller tiles and encoding them using a Variational Autoencoder. #### Ace - [EmptyAceStep1.5LatentAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyAceStep1.5LatentAudio.md): The Empty Ace Step 1.5 Latent Audio node creates an empty latent tensor designed for audio processing. - [EmptyAceStepLatentAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyAceStepLatentAudio.md): The Empty Ace Step 1.0 Latent Audio node creates empty latent audio samples of a specified duration. #### Advanced - [LatentAdd - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentAdd.md): The LatentAdd node is designed for the addition of two latent representations. - [LatentBatchSeedBehavior - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentBatchSeedBehavior.md): The LatentBatchSeedBehavior node is designed to modify the seed behavior of a batch of latent samples. - [LatentConcat - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentConcat.md): The LatentConcat node combines two latent samples by joining them together along a chosen dimension. - [LatentCut - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentCut.md): The LatentCut node extracts a specific section from latent samples along a chosen dimension. - [LatentCutToBatch - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentCutToBatch.md): The LatentCutToBatch node splits a latent representation along a chosen dimension (time, width, or height) into slices of a specified size and stacks them into a new batch. - [LatentInterpolate - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentInterpolate.md): The LatentInterpolate node is designed to perform interpolation between two sets of latent samples based on a specified ratio, blending the characteristics of both sets to produce - [LatentMultiply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentMultiply.md): The LatentMultiply node is designed to scale the latent representation of samples by a specified multiplier. - [LatentSubtract - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentSubtract.md): The LatentSubtract node is designed for subtracting one latent representation from another. ##### Operations - [LatentApplyOperation - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentApplyOperation.md): The LatentApplyOperation node applies a specified latent operation to latent samples. - [LatentApplyOperationCFG - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentApplyOperationCFG.md): The LatentApplyOperationCFG node applies a latent operation to modify the conditioning guidance process in a model. - [LatentOperationSharpen - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentOperationSharpen.md): The LatentOperationSharpen node creates a sharpening operation for latent representations using a Gaussian kernel. - [LatentOperationTonemapReinhard - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentOperationTonemapReinhard.md): The LatentOperationTonemapReinhard node applies Reinhard tonemapping to latent vectors. #### Autoregressive - [EmptyARVideoLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyARVideoLatent.md): Documentation for EmptyARVideoLatent node. #### Batch - [BatchLatentsNode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/BatchLatentsNode.md): The Batch Latents node combines multiple latent inputs into a single batch. - [LatentBatch - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentBatch.md): The LatentBatch node is designed to merge two sets of latent samples into a single batch, potentially resizing one set to match the dimensions of the other before concatenation. - [LatentFromBatch - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentFromBatch.md): This node is designed to extract a specific subset of latent samples from a given batch based on the specified batch index and length. - [RebatchLatents - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/RebatchLatents.md): The RebatchLatents node is designed to reorganize a batch of latent representations into a new batch configuration, based on a specified batch size. - [RepeatLatentBatch - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/RepeatLatentBatch.md): The RepeatLatentBatch node is designed to replicate a given batch of latent representations a specified number of times, potentially including additional data like noise masks and - [ReplaceVideoLatentFrames - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ReplaceVideoLatentFrames.md): ReplaceVideoLatentFrames replaces a range of frames in a destination latent video with frames from a source latent video, starting at a specified frame index. - [SeedVR2TemporalChunk - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SeedVR2TemporalChunk.md): This node splits a SeedVR2 video latent into smaller temporal chunks that can be processed one at a time within available VRAM. - [SeedVR2TemporalMerge - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SeedVR2TemporalMerge.md): This node recombines sampled SeedVR2 latent temporal chunks into a single full-length latent. #### Chroma Radiance - [EmptyChromaRadianceLatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyChromaRadianceLatentImage.md): The EmptyChromaRadianceLatentImage node creates a blank latent image with specified dimensions for use in chroma radiance workflows. #### Cosmos - [EmptyCosmosLatentVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyCosmosLatentVideo.md): EmptyCosmosLatentVideo creates an empty latent video tensor with the specified dimensions. #### Flux - [EmptyFlux2LatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyFlux2LatentImage.md): The Empty Flux 2 Latent node creates a blank, empty latent representation. #### Hidream - [EmptyHiDreamO1LatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyHiDreamO1LatentImage.md): This node creates an empty latent image in pixel space, specifically designed for the HiDream-O1-Image model. #### Hunyhuan Video - [HunyuanVideo15LatentUpscaleWithModel - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HunyuanVideo15LatentUpscaleWithModel.md): The Hunyuan Video 15 Latent Upscale With Model node increases the resolution of a latent image representation. #### Hunyuan 3d - [EmptyLatentHunyuan3Dv2 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyLatentHunyuan3Dv2.md): The EmptyLatentHunyuan3Dv2 node creates blank latent tensors specifically formatted for Hunyuan3Dv2 3D generation models. - [VAEDecodeHunyuan3D - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEDecodeHunyuan3D.md): The VAEDecodeHunyuan3D node converts latent representations into 3D voxel data using a VAE decoder. #### Hunyuan Image - [EmptyHunyuanImageLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyHunyuanImageLatent.md): The EmptyHunyuanImageLatent node creates an empty (zero-filled) latent space for Hunyuan image generation models. #### Hunyuan Video - [EmptyHunyuanLatentVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyHunyuanLatentVideo.md): The EmptyHunyuanLatentVideo node is similar to the EmptyLatentImage node. - [EmptyHunyuanVideo15Latent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyHunyuanVideo15Latent.md): This node creates an empty latent tensor specifically formatted for use with the HunyuanVideo 1.5 model. #### Ltxv - [EmptyLTXVLatentVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyLTXVLatentVideo.md): The EmptyLTXVLatentVideo node creates an empty latent tensor for video processing. - [LTXVAudioVAEDecode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVAudioVAEDecode.md): The LTXV Audio VAE Decode node converts a latent representation of audio back into an audio waveform. - [LTXVAudioVAEEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVAudioVAEEncode.md): The LTXV Audio VAE Encode node takes an audio input and compresses it into a smaller, latent representation using a specified Audio VAE model. - [LTXVConcatAVLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVConcatAVLatent.md): This node merges a video latent and an audio latent into a single joint audio-video (AV) latent, ready for AV models such as LTXV or MiniMax H3. - [LTXVEmptyLatentAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVEmptyLatentAudio.md): markdown - [LTXVLatentUpsampler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVLatentUpsampler.md): The LTXVLatentUpsampler node increases the spatial resolution of a video latent representation by a factor of two. - [LTXVSeparateAVLatent - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVSeparateAVLatent.md): The LTXVSeparateAVLatent node splits a combined audio-visual latent into two separate latents: one containing the video data and one containing the audio data. #### Minimax - [EmptyMiniMaxH3LatentAV - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyMiniMaxH3LatentAV.md): This node creates an empty latent that combines both video and audio information for the MiniMax H3 model. #### Minimax Music - [EmptyMiniMaxMusic3LatentAudio - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyMiniMaxMusic3LatentAudio.md): This node creates an empty (zero-filled) audio latent for the MiniMax Music3 model. #### Mochi - [EmptyMochiLatentVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyMochiLatentVideo.md): EmptyMochiLatentVideo creates an empty latent video tensor with the dimensions you specify. #### Qwen - [EmptyQwenImageLayeredLatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyQwenImageLayeredLatentImage.md): Empty Qwen Image Layered Latent prepares the blank canvas that the Qwen-Image-Layered model paints onto. #### Stable Cascade - [StableCascade_EmptyLatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StableCascade_EmptyLatentImage.md): The StableCascadeEmptyLatentImage node creates empty latent tensors for Stable Cascade models. - [StableCascade_StageC_VAEEncode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StableCascade_StageC_VAEEncode.md): The StableCascadeStageCVAEEncode node processes an input image through a VAE encoder to generate latent representations for the Stable Cascade model. #### Stable Diffusion - [EmptySD3LatentImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptySD3LatentImage.md): EmptySD3LatentImage creates a blank latent image tensor specifically formatted for Stable Diffusion 3 models. #### Transform - [LatentCrop - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentCrop.md): The LatentCrop node is designed to perform cropping operations on latent representations of images. - [LatentFlip - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentFlip.md): The LatentFlip node is designed to manipulate latent representations by flipping them either vertically or horizontally. - [LatentRotate - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentRotate.md): The LatentRotate node is designed to rotate latent representations of images by specified angles. #### Trellis - [EmptyTrellis2LatentStructure - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/EmptyTrellis2LatentStructure.md): This node creates an empty latent structure for the Trellis2 model, where all values are set to zero. - [VaeDecodeShapeTrellis - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VaeDecodeShapeTrellis.md): This node decodes Trellis2 shape latent representations into a 3D mesh. - [VaeDecodeStructureTrellis2 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VaeDecodeStructureTrellis2.md): This node converts Trellis structure latent samples into a 3D voxel grid using a VAE's structure decoder. - [VaeDecodeTextureTrellis - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VaeDecodeTextureTrellis.md): This node decodes a Trellis2 texture latent into voxel colors using a VAE. #### Triposplat - [TripoSplatSamplingPreview - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TripoSplatSamplingPreview.md): This node patches a TripoSplat model so that when used with the standard KSampler node, a live preview of the decoded gaussian splat is shown at each sampling step. - [VAEDecodeTripoSplat - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAEDecodeTripoSplat.md): Decode a TripoSplat latent representation into a 3D gaussian splat. #### Void - [VOIDWarpedNoise - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VOIDWarpedNoise.md): Generates temporally-correlated noise for the second pass of the VOID video refinement process. - [VOIDWarpedNoiseSource - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VOIDWarpedNoiseSource.md): Documentation for VOIDWarpedNoiseSource node. ### Loaders - [AudioEncoderLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/AudioEncoderLoader.md): The AudioEncoderLoader node loads an audio encoder model from a file in your audio encoders folder. - [CheckpointLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CheckpointLoader.md): The CheckpointLoader node loads a pre-trained model checkpoint together with its configuration file. - [CheckpointLoaderSimple - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CheckpointLoaderSimple.md): Loads a diffusion model checkpoint file and decomposes it into three core components: the main model used for denoising latents, the CLIP text encoder, and the VAE image encoder/de - [CLIPLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipLoader.md): The CLIPLoader node loads a text encoder model (CLIP, T5, or similar) from a file, making it available for use in other nodes that need to convert text prompts into numerical repre - [CLIPVisionLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipVisionLoader.md): This node automatically detects models located in the ComfyUI/models/clipvision folder, as well as any additional model paths configured in the extramodelpaths.yaml file. - [DiffControlNetLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/DiffControlNetLoader.md): This node will detect models located in the ComfyUI/models/controlnet folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [DiffusersLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/DiffusersLoader.md): The DiffusersLoader node loads pre-trained models saved in the diffusers format. - [DualCLIPLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/DualCLIPLoader.md): The DualCLIPLoader node is designed for loading two CLIP models simultaneously, facilitating operations that require the integration or comparison of features from both models. - [FrameInterpolationModelLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FrameInterpolationModelLoader.md): Documentation for FrameInterpolationModelLoader node. - [GLIGENLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/GLIGENLoader.md): This node will detect models located in the ComfyUI/models/gligen folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [HypernetworkLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HypernetworkLoader.md): This node will detect models located in the ComfyUI/models/hypernetworks folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [ImageOnlyCheckpointLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ImageOnlyCheckpointLoader.md): This node will detect models located in the ComfyUI/models/checkpoints folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [LatentUpscaleModelLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LatentUpscaleModelLoader.md): The LatentUpscaleModelLoader node loads a model specialized in upscaling latent representations from a file stored in ComfyUI's latentupscalemodels folder. - [LoadBackgroundRemovalModel - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoadBackgroundRemovalModel.md): Loads a background removal model from a file. - [LoadDA3Model - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoadDA3Model.md): The Load Depth Anything 3 node loads a Depth Anything 3 model from a file, preparing it for depth estimation tasks. - [LoadMediaPipeFaceLandmarker - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoadMediaPipeFaceLandmarker.md): Load Face Detection Model (MediaPipe) - [LoadMoGeModel - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoadMoGeModel.md): Loads a MoGe (Monocular Geometry) model from a file and prepares it for use in geometry estimation tasks. - [LoraLoaderBypass - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoraLoaderBypass.md): The LoraLoaderBypass node applies a LoRA (Low-Rank Adaptation) to a diffusion model and a CLIP model in a special "bypass" mode. - [LoraLoaderBypassModelOnly - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoraLoaderBypassModelOnly.md): This node applies a LoRA (Low-Rank Adaptation) to a model to modify its behavior, but only affects the model component itself. - [LoraModelLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoraModelLoader.md): The LoraModelLoader node applies trained LoRA (Low-Rank Adaptation) weights to a diffusion model. - [LTXAVTextEncoderLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXAVTextEncoderLoader.md): This node loads a specialized text encoder for the LTXV audio model. - [LTXVAudioVAELoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVAudioVAELoader.md): The LTXV Audio VAE Loader node loads a pre-trained Audio Variational Autoencoder (VAE) model from a checkpoint file. - [ModelPatchLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelPatchLoader.md): The ModelPatchLoader node loads a model patch file from the modelpatches folder and prepares it for use in a workflow. - [OpticalFlowLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/OpticalFlowLoader.md): Loads an optical flow model from the models/opticalflow/ folder. - [PhotoMakerLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/PhotoMakerLoader.md): The PhotoMakerLoader node loads a PhotoMaker model from the available model files. - [QuadrupleCLIPLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/QuadrupleCLIPLoader.md): The Quadruple CLIP Loader, QuadrupleCLIPLoader, is one of the core nodes of ComfyUI, first added to support the HiDream I1 version model. - [StyleModelLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/StyleModelLoader.md): This node will detect models located in the ComfyUI/models/stylemodels folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [TripleCLIPLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TripleCLIPLoader.md): TripleCLIPLoader loads three text encoder models at the same time and combines them into a single CLIP model. - [unCLIPCheckpointLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/unCLIPCheckpointLoader.md): This node will detect models located in the ComfyUI/models/checkpoints folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [UNETLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/UNETLoader.md): The UNETLoader node is designed for loading U-Net models by name, facilitating the use of pre-trained U-Net architectures within the system. - [UpscaleModelLoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/UpscaleModelLoader.md): This node will detect models located in the ComfyUI/models/upscalemodels folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. - [VAELoader - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAELoader.md): This node will detect models located in the ComfyUI/models/vae folder, and it will also read models from additional paths configured in the extramodelpaths.yaml file. ### Merging - [CheckpointSave - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CheckpointSave.md): The Save Checkpoint node is designed to save a complete Stable Diffusion model (including UNet, CLIP, and VAE components) as a .safetensors format checkpoint file. - [CLIPMergeAdd - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPMergeAdd.md): The CLIPMergeAdd node combines two CLIP models by adding patches from the second model to the first model. - [CLIPMergeSimple - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipMergeSimple.md): CLIPMergeSimple is a model merging node that combines two CLIP text encoder models based on a specified ratio. - [CLIPMergeSubtract - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CLIPMergeSubtract.md): The CLIPMergeSubtract node performs model merging by subtracting the weights of one CLIP model from another. - [CLIPSave - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ClipSave.md): The CLIPSave node saves a CLIP text encoder model to disk in SafeTensors format. - [ImageOnlyCheckpointSave - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ImageOnlyCheckpointSave.md): The ImageOnlyCheckpointSave node saves a checkpoint file containing a model, CLIP vision encoder, and VAE. - [ModelMergeAdd - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeAdd.md): The ModelMergeAdd node is designed for merging two models by adding key patches from one model to another. - [ModelMergeBlocks - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeBlocks.md): ModelMergeBlocks is designed for advanced model merging operations, allowing for the integration of two models with customizable blending ratios for different parts of the models. - [ModelMergeSimple - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeSimple.md): The ModelMergeSimple node is designed for merging two models by blending their parameters based on a specified ratio. - [ModelMergeSubtract - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeSubtract.md): This node is designed for advanced model merging operations, specifically to subtract the parameters of one model from another based on a specified multiplier. - [ModelSave - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSave.md): The ModelSave node saves a model to your computer's storage as a .safetensors checkpoint file. - [SaveLoRA - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SaveLoRA.md): The SaveLoRA node saves a LoRA (Low-Rank Adaptation) model to a file. - [VAESave - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VAESave.md): The VAESave node is designed for saving VAE models along with their metadata, including prompts and additional PNG information, to a specified output directory. #### Model Specific - [ModelMergeAuraflow - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeAuraflow.md): The ModelMergeAuraflow node allows you to blend two different models together by adjusting specific blending weights for various model components. - [ModelMergeCosmos14B - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeCosmos14B.md): The ModelMergeCosmos14B node merges two AI models using a block-based approach specifically designed for the Cosmos 14B model architecture. - [ModelMergeCosmos7B - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeCosmos7B.md): The ModelMergeCosmos7B node merges two AI models together using weighted blending of specific components. - [ModelMergeCosmosPredict2_14B - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeCosmosPredict2_14B.md): The ModelMergeCosmosPredict214B node merges two AI models by blending their internal components. - [ModelMergeCosmosPredict2_2B - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeCosmosPredict2_2B.md): The ModelMergeCosmosPredict22B node merges two diffusion models using a block-based approach with fine-grained control over different model components. - [ModelMergeFlux1 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeFlux1.md): The ModelMergeFlux1 node merges two diffusion models by blending their components using weighted interpolation. - [ModelMergeKrea2 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeKrea2.md): This node merges two models by blending their internal components at a fine-grained level, allowing you to control how much of each model's specific parts influence the final resul - [ModelMergeLTXV - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeLTXV.md): The ModelMergeLTXV node performs advanced model merging operations specifically designed for LTXV model architectures. - [ModelMergeMochiPreview - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeMochiPreview.md): This node merges two Mochi AI models using a block-based approach with fine-grained control over different model components. - [ModelMergeQwenImage - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeQwenImage.md): ModelMergeQwenImage merges two AI models by combining their components with adjustable weights. - [ModelMergeSD1 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeSD1.md): The ModelMergeSD1 node blends two Stable Diffusion 1.x models together by adjusting how much each model component contributes to the result. - [ModelMergeSD35_Large - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeSD35_Large.md): The ModelMergeSD35Large node merges two Stable Diffusion 3.5 Large models by blending specific internal components of the second model into the first. - [ModelMergeSD3_2B - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeSD3_2B.md): The ModelMergeSD32B node allows you to merge two Stable Diffusion 3 2B models by blending their components with adjustable weights. - [ModelMergeSDXL - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeSDXL.md): The ModelMergeSDXL node allows you to blend two SDXL models together by adjusting the influence of each model on different parts of the architecture. - [ModelMergeWAN2_1 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelMergeWAN2_1.md): The ModelMergeWAN21 node merges two WAN2.1 models by blending their components using weighted averages. ### Patch - [ContextWindowsManual - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ContextWindowsManual.md): The Context Windows (Manual) node allows you to manually configure context windows for a model during sampling, creating overlapping context segments with a specified length, overl - [ModelAttentionBackend - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelAttentionBackend.md): This node lets you choose which attention backend a model uses for its attention computations. - [ModelNoiseScale - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelNoiseScale.md): Documentation for ModelNoiseScale node. - [ModelSamplingAuraFlow - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingAuraFlow.md): The ModelSamplingAuraFlow node applies a specialized sampling configuration to diffusion models, specifically designed for AuraFlow model architectures. - [ModelSamplingContinuousEDM - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingContinuousEDM.md): This node is designed to enhance a model's sampling capabilities by integrating continuous EDM (Energy-based Diffusion Models) sampling techniques. - [ModelSamplingContinuousV - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingContinuousV.md): The ModelSamplingContinuousV node adjusts a model's sampling behavior by applying continuous V-prediction sampling. - [ModelSamplingDiscrete - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingDiscrete.md): This node is designed to modify the sampling behavior of a model by applying a discrete sampling strategy. - [RenormCFG - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/RenormCFG.md): The RenormCFG node modifies the classifier-free guidance (CFG) process in diffusion models by applying conditional scaling and normalization. - [RescaleCFG - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/RescaleCFG.md): The RescaleCFG node is designed to adjust the conditioning and unconditioning scales of a model's output based on a specified multiplier, aiming to achieve a more balanced and cont - [ScaleROPE - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ScaleROPE.md): The ScaleROPE node allows you to modify the Rotary Position Embedding (ROPE) of a model by applying separate scaling and shifting factors to its X, Y, and T (time) components. #### Anima - [AnimaLLLiteApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/AnimaLLLiteApply.md): AnimaLLLiteApply applies a lightweight animation patch to a diffusion model, enabling controlled image-to-image generation with adjustable strength and timing. #### Chroma Radiance - [ChromaRadianceOptions - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ChromaRadianceOptions.md): The ChromaRadianceOptions node allows you to configure advanced settings for the Chroma Radiance model. #### Flux - [ModelSamplingFlux - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingFlux.md): The ModelSamplingFlux node applies Flux model sampling to a given model by calculating a shift parameter based on image dimensions. - [USOStyleReference - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/USOStyleReference.md): The USOStyleReference node applies a style reference to a model by combining CLIP vision features with a model patch, and returns a patched copy of the input model. #### Hidream - [HiDreamO1PatchSeamSmoothing - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HiDreamO1PatchSeamSmoothing.md): This node reduces visible seams in images generated by the HiDream-O1 model by averaging the model's output across multiple shifted patch-grid positions during the later part of th #### Ltxv - [LTXVContextWindows - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVContextWindows.md): This node sets context windows for LTXV-like models during sampling. - [ModelSamplingLTXV - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingLTXV.md): The ModelSamplingLTXV node applies advanced sampling parameters to a model based on token count. #### Minimax - [MiniMaxH3FunControlNetApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MiniMaxH3FunControlNetApply.md): This node applies a MiniMax H3 Fun ControlNet to a text-to-video model as a model patch. - [MiniMaxH3SigmaShift - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MiniMaxH3SigmaShift.md): Sets the video and audio flow shift values for a MiniMax H3 model. #### Qwen - [QwenImageDiffsynthControlnet - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/QwenImageDiffsynthControlnet.md): QwenImageDiffsynthControlnet applies a diffusion synthesis control network patch to a base model. #### Sensenova - [SenseNovaSamplingOptions - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SenseNovaSamplingOptions.md): SenseNova Sampling Options sets the SenseNova flow shift on a model. #### Stable Cascade - [ModelSamplingStableCascade - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingStableCascade.md): The ModelSamplingStableCascade node applies stable cascade sampling to a model by adjusting the sampling parameters with a shift value. #### Stable Diffusion - [ModelSamplingSD3 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ModelSamplingSD3.md): The ModelSamplingSD3 node applies Stable Diffusion 3 sampling parameters to a model. #### Supir - [SUPIRApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SUPIRApply.md): The SUPIRApply node applies a SUPIR model patch to a diffusion model. #### Unet - [Epsilon Scaling - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Epsilon Scaling.md): This node implements the Epsilon Scaling method from the research paper "Elucidating the Exposure Bias in Diffusion Models" (arxiv.org/abs/2308.15321v6). - [FreeU - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FreeU.md): The FreeU node applies frequency-domain modifications to a model's output blocks to enhance image generation quality. - [FreeU_V2 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FreeU_V2.md): FreeUV2 enhances image generation quality by applying frequency-based modifications to a diffusion model's U-Net architecture. - [HyperTile - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/HyperTile.md): The HyperTile node applies a tiling technique to the attention mechanism in diffusion models to optimize memory usage during image generation. - [PatchModelAddDownscale - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/PatchModelAddDownscale.md): PatchModelAddDownscale (Kohya Deep Shrink) implements the Kohya Deep Shrink technique by applying downscaling and upscaling operations to specific blocks in a model. - [PerturbedAttentionGuidance - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/PerturbedAttentionGuidance.md): The PerturbedAttentionGuidance node applies perturbed attention guidance to a diffusion model to enhance generation quality. - [TemporalScoreRescaling - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TemporalScoreRescaling.md): This node applies Temporal Score Rescaling (TSR) to a diffusion model. - [TomePatchModel - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TomePatchModel.md): TomePatchModel applies Token Merging (ToMe) to a diffusion model to reduce computational requirements during inference. #### Wan - [WanContextWindowsManual - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanContextWindowsManual.md): The WAN Context Windows (Manual) node lets you manually configure context windows for Wan-style video models. - [WanUni3CControlnetApply - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/WanUni3CControlnetApply.md): Documentation for WanUni3CControlnetApply node. #### Z Image - [ZImageFunControlnet - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ZImageFunControlnet.md): ZImageFunControlnet applies a specialized control network to influence the image generation or editing process. ### Sampling - [KSampler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/KSampler.md): The KSampler works like this: it modifies the provided original latent image information based on a specific model and both positive and negative conditions. - [KSamplerAdvanced - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/KSamplerAdvanced.md): The KSamplerAdvanced node is designed to enhance the sampling process by providing advanced configurations and techniques. #### Custom - [APG - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/APG.md): The APG (Adaptive Projected Guidance) node modifies the sampling process by adjusting how guidance is applied during diffusion. - [SamplerCustom - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerCustom.md): The SamplerCustom node is designed to provide a flexible and customizable sampling mechanism for various applications. - [SamplerCustomAdvanced - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerCustomAdvanced.md): The SamplerCustomAdvanced node performs advanced latent space sampling using custom noise, guidance, and sampling configurations. #### Guiders - [BasicGuider - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/BasicGuider.md): The BasicGuider node creates a simple guidance mechanism for the sampling process. - [CFGGuider - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CFGGuider.md): The CFG Guider node creates a guidance system for controlling the sampling process in image generation. - [CFGOverride - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/CFGOverride.md): The CFG Override node lets you set a fixed CFG (Classifier-Free Guidance) scale value for a specific range of the sampling process, defined as a percentage of the total steps. - [DualCFGGuider - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/DualCFGGuider.md): The DualCFGGuider node creates a guidance system for dual classifier-free guidance sampling. - [DualModelGuider - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/DualModelGuider.md): This node allows you to use two different models during the guided CFG sampling process: one model for the positive (conditional) pass and a separate model for the negative (uncond - [LTXVDualCFGGuider - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVDualCFGGuider.md): This node creates a guided sampling object (CFG guider) for LTXV-AV models. - [VideoLinearCFGGuidance - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VideoLinearCFGGuidance.md): The VideoLinearCFGGuidance node applies a linear conditioning guidance scale to a video model, adjusting the influence of conditioned and unconditioned components over a specified - [VideoTriangleCFGGuidance - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VideoTriangleCFGGuidance.md): The VideoTriangleCFGGuidance node applies a triangular classifier-free guidance (CFG) scaling pattern to a video model. #### Noise - [AddNoise - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/AddNoise.md): This node adds controlled noise to a latent image using a specified noise generator and sigma values. - [DisableNoise - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/DisableNoise.md): The DisableNoise node provides an empty noise configuration that can be used to disable noise generation in sampling processes. - [RandomNoise - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/RandomNoise.md): The RandomNoise node creates a noise generator based on a seed value for use during the sampling process. #### Samplers - [KSamplerSelect - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/KSamplerSelect.md): The KSamplerSelect node is designed to select a specific sampler based on the provided sampler name. - [SamplerARVideo - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerARVideo.md): The Sampler AR Video node provides a specialized sampling method for autoregressive video models, such as those using Causal Forcing or Self-Forcing techniques. - [SamplerDPMAdaptative - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerDPMAdaptative.md): The SamplerDPMAdaptative node implements an adaptive DPM (Diffusion Probabilistic Model) sampler that automatically adjusts step sizes during the sampling process. - [SamplerDPMPP_2M_SDE - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerDPMPP_2M_SDE.md): The SamplerDPMPP2MSDE node creates a DPM++ 2M SDE sampler for diffusion models. - [SamplerDPMPP_2S_Ancestral - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerDPMPP_2S_Ancestral.md): The SamplerDPMPP2SAncestral node creates a sampler that uses the DPM++ 2S Ancestral sampling method for generating images. - [SamplerDPMPP_3M_SDE - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerDPMPP_3M_SDE.md): The SamplerDPMPP3MSDE node creates a DPM++ 3M SDE sampler for use in the sampling process. - [SamplerDPMPP_SDE - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerDPMPP_SDE.md): SamplerDPMPPSDE creates a DPM++ SDE (Stochastic Differential Equation) sampler for use in the sampling process. - [SamplerER_SDE - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerER_SDE.md): The SamplerERSDE node provides specialized sampling methods for diffusion models, offering different solver types: ER-SDE, Reverse-time SDE, and ODE. - [SamplerEulerAncestral - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerEulerAncestral.md): The SamplerEulerAncestral node creates an Euler Ancestral sampler for generating images. - [SamplerEulerAncestralCFGPP - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerEulerAncestralCFGPP.md): The SamplerEulerAncestralCFGPP node creates a sampler that uses the Euler Ancestral method with classifier-free guidance (CFG++) for image generation. - [SamplerLCM - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerLCM.md): The SamplerLCM node provides an LCM (Latent Consistency Model) sampler with tunable per-step noise parameters. - [SamplerLCMUpscale - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerLCMUpscale.md): The SamplerLCMUpscale node provides a specialized sampling method that combines Latent Consistency Model (LCM) sampling with image upscaling capabilities. - [SamplerLMS - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerLMS.md): The SamplerLMS node creates a Least Mean Squares (LMS) sampler for use in diffusion models. - [SamplerSASolver - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerSASolver.md): The SamplerSASolver node implements a custom sampling algorithm for diffusion models. - [SamplerSEEDS2 - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplerSEEDS2.md): This node provides a configurable sampler for image generation. - [VOIDSampler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VOIDSampler.md): VOIDSampler is a specialized DDIM sampler for VOID inpainting models. #### Schedulers - [AlignYourStepsScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/AlignYourStepsScheduler.md): The AlignYourStepsScheduler node generates sigma values for the denoising process based on different model types. - [BasicScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/BasicScheduler.md): The BasicScheduler node is designed to compute a sequence of sigma values for diffusion models based on the provided scheduler, model, and denoising parameters. - [BetaSamplingScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/BetaSamplingScheduler.md): The BetaSamplingScheduler node generates a sequence of noise levels (sigmas) for the sampling process using a beta scheduling algorithm. - [ExponentialScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ExponentialScheduler.md): The ExponentialScheduler node is designed to generate a sequence of sigma values following an exponential schedule for diffusion sampling processes. - [Flux2Scheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Flux2Scheduler.md): Flux2Scheduler generates a sequence of noise levels (sigmas) for the denoising process, specifically tailored for the Flux model. - [GITSScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/GITSScheduler.md): The GITSScheduler node generates noise schedule sigmas for the GITS (Generative Iterative Time Steps) sampling method. - [Ideogram4Scheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/Ideogram4Scheduler.md): The Ideogram 4 Scheduler node generates a sequence of sigma values (noise levels) for the diffusion sampling process, based on the Ideogram 4 reference schedule. - [KarrasScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/KarrasScheduler.md): The KarrasScheduler node is designed to generate a sequence of noise levels (sigmas) based on the Karras et al. (2022) noise schedule. - [LaplaceScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LaplaceScheduler.md): The LaplaceScheduler node generates a sequence of sigma values following a Laplace distribution for use in diffusion sampling. - [LTXVScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LTXVScheduler.md): The LTXVScheduler node generates sigma values for custom sampling processes. - [OptimalStepsScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/OptimalStepsScheduler.md): The OptimalStepsScheduler node creates a noise schedule (a sequence of sigma values) for use during diffusion sampling. - [PolyexponentialScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/PolyexponentialScheduler.md): The PolyexponentialScheduler node is designed to generate a sequence of noise levels (sigmas) based on a polyexponential noise schedule. - [SDTurboScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SDTurboScheduler.md): SDTurboScheduler is designed to generate a sequence of sigma values for image sampling, adjusting the sequence based on the denoise level and the number of steps specified. - [VPScheduler - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/VPScheduler.md): The VPScheduler node is designed to generate a sequence of noise levels (sigmas) based on the Variance Preserving (VP) scheduling method. #### Sigmas - [ExtendIntermediateSigmas - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ExtendIntermediateSigmas.md): The ExtendIntermediateSigmas node takes an existing sequence of sigma values and inserts additional intermediate sigma values between them. - [FlipSigmas - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/FlipSigmas.md): The FlipSigmas node is designed to manipulate the sequence of sigma values used in diffusion models by reversing their order and ensuring the first value is non-zero if originally - [ManualSigmas - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ManualSigmas.md): The ManualSigmas node allows you to manually define a custom sequence of noise levels (sigmas) for the sampling process. - [SamplingPercentToSigma - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SamplingPercentToSigma.md): The SamplingPercentToSigma node converts a sampling percentage value to a corresponding sigma value using the model's sampling parameters. - [SetFirstSigma - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SetFirstSigma.md): The SetFirstSigma node changes a sigma sequence by replacing only its first value with a custom sigma value. - [SplitSigmas - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SplitSigmas.md): The SplitSigmas node is designed for dividing a sequence of sigma values into two parts based on a specified step. - [SplitSigmasDenoise - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SplitSigmasDenoise.md): The SplitSigmasDenoise node divides a sequence of sigma values into two parts based on a denoising strength parameter. ### Training - [LoadTrainingDataset - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LoadTrainingDataset.md): This node loads an encoded training dataset (latents and conditioning) that was previously saved to disk. - [LossGraphNode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/LossGraphNode.md): The LossGraphNode creates a line chart of training loss values over training steps and shows it as a preview image. - [MakeTrainingDataset - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/MakeTrainingDataset.md): This node prepares data for training by encoding images and text. - [ResolutionBucket - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/ResolutionBucket.md): This node organizes a list of latent images and their corresponding conditioning data by their resolution. - [SaveTrainingDataset - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/SaveTrainingDataset.md): This node saves an encoded training dataset to disk for efficient loading during training. - [TrainLoraNode - ComfyUI Built-in Node Documentation](https://docs.comfy.org/built-in-nodes/TrainLoraNode.md): The TrainLoraNode creates and trains a LoRA (Low-Rank Adaptation) model on a diffusion model using provided latents and conditioning data.