Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
Generate a prompt-directed video variation from an uploaded clip using Capybara. Selected source frames are resized and supplied to video conditioning and, by default, a visual encoder; the generated frames become the output video.
Workflow details
Runtime profile
Sources
1 sourceHover nodes for Pipe controls
Core
Primary Pipe surface
Source video
LoadVideo
Edit instruction
CLIPTextEncode
Output width
PrimitiveInt
Output height
PrimitiveInt
Generated frames
HunyuanVideo15ImageToVideo
Playback and bit depth
CreateVideo
Extended
Additional surfaced controls
Negative prompt
CLIPTextEncode
Source frame range
ImageFromBatch
Source framing
ResizeImageMaskNode
Visual guidance
CLIPVisionEncode
Randomization
RandomNoise
Prompt guidance
CFGGuider
Sampling
BasicScheduler
Sampler
KSamplerSelect
Sampling shift
ModelSamplingSD3
Video output
SweetTeaPreviewVideo
Hardware and download-size estimates are not available yet.
Requirements
byt5_small_glyphxl_fp16.safetensors
Not resolvedText encoder · Hugging Face · Comfy-Org/HunyuanVideo_1.5_repackaged
Checkpoint · Hugging Face · Comfy-Org/HunyuanVideo_1.5_repackaged
hunyuanvideo15_vae_fp16.safetensors
Not resolvedVAE · Hugging Face · Comfy-Org/HunyuanVideo_1.5_repackaged
Text encoder · Hugging Face · Comfy-Org/HunyuanImage_2.1_ComfyUI · HunyuanImage 2.1
sigclip_vision_patch14_384.safetensors
File unverifiedVision model · Hugging Face · Comfy-Org/sigclip_vision_384 · Sigclip Vision 384
Node pack · GitHub · tea-time-labs/sweet-tea-nodes
