floyo logo
Workflows
Pricing
floyo logo
Workflows
Pricing
Last updated
2026-05-11

LTX-Video Support for ComfyUI is an extension that enhances the capabilities of the ComfyUI interface by integrating advanced video generation features from the LTX-2 model. This tool provides a collection of custom nodes that allow users to create and manipulate video content efficiently.

  • Offers a variety of workflows for transforming text and images into videos, as well as enhancing existing video content.
  • Introduces specialized models like Union IC-LoRA and HDR IC-LoRA for advanced control over video attributes and quality.
  • Supports multilingual dubbing and rephrasing of audio in videos, maintaining speaker identity and coherence.

Context

This tool serves as an extension to ComfyUI, focusing on integrating the LTX-2 video generation model, which is embedded in the core of ComfyUI. It aims to provide users with additional nodes and workflows that enhance the functionality of LTX-2, enabling more sophisticated video generation and editing capabilities.

Key Features & Benefits

The primary features of this tool include various workflows that facilitate the conversion of text and images into videos, as well as the ability to upscale and refine video content. The inclusion of unique models such as the Union IC-LoRA allows for efficient processing with multiple control signals, while the HDR IC-LoRA enables the output of high-dynamic-range video, which is crucial for professional video production.

Advanced Functionalities

Among its advanced capabilities, the tool features a Union IC-LoRA model that combines depth and edge control conditions into a single LoRA, enhancing performance and efficiency. Additionally, the HDR IC-LoRA model outputs linear HDR video suitable for grading, allowing for more dynamic visual content. The Lipdub IC-LoRA enables real-time dubbing and rephrasing of dialogue in videos, supporting multilingual capabilities while preserving the original speaker's identity.

Practical Benefits

This extension significantly enhances the workflow in ComfyUI by providing users with greater control over video generation processes, improving the quality of outputs, and streamlining the integration of various video attributes. The advanced models and workflows not only increase efficiency but also allow for high-quality video production that meets professional standards.

Credits/Acknowledgments

The LTX-Video Support for ComfyUI is developed by Lightricks, with contributions from various collaborators. The repository is open-source, and users are encouraged to refer to the original GitHub repository for more detailed information and resources.

Inner Nodes

AddLatentGuide
DynamicConditioning
GemmaAPITextEncode
GuiderParameters
ImageToCPU
LTXAddImageGuide
LTXAddVideoICLoRAGuide
LTXAddVideoICLoRAGuideAdvanced
LTXAttentioOverride
LTXAttentionBank
LTXAttnOverride
LTXFetaEnhance
LTXFloatToInt
LTXFlowEditCFGGuider
LTXFlowEditSampler
LTXForwardModelSamplingPred
LTXICLoRALoaderModelOnly
LTXPerturbedAttention
LTXPrepareAttnInjections
LTXQ8Patch
LTXRFForwardODESampler
LTXRFReverseODESampler
LTXReverseModelSamplingPred
LTXVAdainLatent
LTXVAddGuideAdvanced
LTXVAddGuideAdvancedAttention
LTXVAddLatentGuide
LTXVApplySTG
LTXVBaseSampler
LTXVDilateVideoMask
LTXVDrawTracks
LTXVExtendSampler
LTXVGemmaCLIPModelLoader
LTXVGemmaEnhancePrompt
LTXVHDRDecodePostprocess
LTXVImgToVideoConditionOnly
LTXVInContextSampler
LTXVInpaintPreprocess
LTXVLaplacianPyramidBlend
LTXVLinearOverlapLatentTransition
LTXVLoadConditioning
LTXVLoopingSampler
LTXVMultiPromptProvider
LTXVNormalizingSampler
LTXVPatcherVAE
LTXVPerStepAdainPatcher
LTXVPerStepStatNormPatcher
LTXVPreprocessMasks
LTXVPromptEnhancer
LTXVPromptEnhancerLoader
LTXVQ8LoraModelLoader
LTXVSaveConditioning
LTXVSelectLatents
LTXVSetAudioRefTokens
LTXVSetVideoLatentNoiseMasks
LTXVSparseTrackEditor
LTXVStatNormLatent
LTXVTiledSampler
LTXVTiledVAEDecode
LowVRAMAudioVAELoader
LowVRAMCheckpointLoader
LowVRAMLatentUpscaleModelLoader
ModifyLTXModel
MultimodalGuider
STGAdvancedPresets
STGGuiderAdvanced
STGGuiderNode
Set VAE Decoder Noise