floyo logo
Powered by
ThinkDiffusion
⚡️Nano Banana 2 ⚡️ just landed. Start creating now.
floyo logo
Powered by
ThinkDiffusion
⚡️Nano Banana 2 ⚡️ just landed. Start creating now.

Wan2.1 InfiniteTalk Video to Video

504

Generates in about -- secs

Nodes & Models

GetNode
CLIPVisionLoader
clip_vision_h.safetensors
LoadAudio
MarkdownNote
Note
SetNode
PreviewAny
INTConstant
INTConstant
ImageResizeKJv2
GetImageRangeFromBatch
GetImageSizeAndCount
ImageConcatMulti
WanVideoLoraSelect
lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16.safetensors
WanVideoVAELoader
Wan2_1_VAE_bf16.safetensors
WanVideoBlockSwap
WanVideoTorchCompileSettings
WanVideoModelLoader
wan2.1_i2v_480p_14B_fp16.safetensors
WanVideoLoraSelect
lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16.safetensors
MultiTalkModelLoader
Wan2_1-InfiniTetalk-Single_fp16.safetensors
WanVideoVAELoader
Wan2_1_VAE_bf16.safetensors
WanVideoBlockSwap
WanVideoTorchCompileSettings
DownloadAndLoadWav2VecModel
WanVideoTextEncodeCached
umt5-xxl-enc-bf16.safetensors
WanVideoModelLoader
wan2.1_i2v_480p_14B_fp16.safetensors
WanVideoApplyNAG
WanVideoEncode
MultiTalkWav2VecEmbeds
WanVideoClipVisionEncode
WanVideoImageToVideoMultiTalk
WanVideoSampler
WanVideoDecode
MultiTalkModelLoader
Wan2_1-InfiniTetalk-Single_fp16.safetensors
DownloadAndLoadWav2VecModel
MultiTalkWav2VecEmbeds
WanVideoImageToVideoMultiTalk
DownloadAndLoadWav2VecModel
VHS_LoadVideo
VHS_VideoCombine
AudioCrop
AudioSeparation
ImageResizeKJv2
GetImageRangeFromBatch
GetImageSizeAndCount
ImageConcatMulti
AudioSeparation
WanVideoEncode

Overview

Wan2.1 InfiniteTalk Video-to-Video enables unlimited-length, audio-driven talking video generation that seamlessly synchronizes lip movement, head turns, body posture, and facial expressions to new audio, all while preserving identity and providing stable, natural animation at up to 720p resolution.

Key Inputs

  1. Audio - High quality audio file

  2. Video - High quality video which displays a single character in the video

  3. A simple detailed prompt

Read more

N