Skip to main content
The ImageOnlyCheckpointSave node saves a checkpoint file containing a model, CLIP vision encoder, and VAE. It creates a safetensors file with the specified filename prefix and stores it in the output directory. This node is specifically designed for saving image-related model components together in a single checkpoint file.

Inputs

ParameterDescriptionData TypeRequiredRange
modelThe model to be saved in the checkpointMODELYes-
clip_visionThe CLIP vision encoder to be saved in the checkpointCLIP_VISIONYes-
vaeThe VAE (Variational Autoencoder) to be saved in the checkpointVAEYes-
filename_prefixThe prefix for the output filename (default: “checkpoints/ComfyUI”)STRINGYes-
promptHidden parameter for workflow prompt dataPROMPTNo-
extra_pnginfoHidden parameter for additional PNG metadataEXTRA_PNGINFONo-

Outputs

Output NameDescriptionData Type
-This node does not return any outputs-
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): d2a26933f0e2fcccf3c57f50038fb40ef5b23d00ccdd2e1d215b3cb78203b9fd