https://github.com/zer0int/comfyui-workflows
Workflows to implement fine-tuned CLIP Text Encoders with ComfyUI / SD, SDXL, SD3
https://github.com/zer0int/comfyui-workflows
Last synced: 3 months ago
JSON representation
Workflows to implement fine-tuned CLIP Text Encoders with ComfyUI / SD, SDXL, SD3
- Host: GitHub
- URL: https://github.com/zer0int/comfyui-workflows
- Owner: zer0int
- Created: 2024-06-19T06:08:50.000Z (11 months ago)
- Default Branch: CLIP-vision
- Last Pushed: 2024-10-08T16:22:24.000Z (8 months ago)
- Last Synced: 2025-01-02T17:33:08.923Z (5 months ago)
- Size: 24.4 KB
- Stars: 3
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# ComfyUI-workflows
Workflows to implement fine-tuned CLIP Text Encoders with ComfyUI / SD, SDXL, SD3, Flux
----
## Update 10/OCT/2024:
- Added `node-a-good-idea.json`- Is basically bonkers and node-a-good-idea to use
- Requires [my Flux Layer Shuffle nodes](https://github.com/zer0int/ComfyUI-CLIP-Flux-Layer-Shuffle)
- Requires [my fork of Block Patcher](https://github.com/zer0int/Block_Patcher_ComfyUI), unless my "custom sigmas" pull is accepted
- Maximizes the ways in which the models involved in Flux can be manipulated in incomprehensible ways
- Use at your own discretion
- I am sorry for the confusion! /o\
----
- 📄 ComfyUI-SDXL-save-and-load-custom-TE-CLIP-finetune.json
- Simple workflow to add e.g. [my custom fine-tuned CLIP ViT-L TE](https://huggingface.co/zer0int/CLIP-GmP-ViT-L-14/blob/main/ViT-L-14-GmP-ft-TE-only-HF-format.safetensors) to SDXL.
----
more soon