README.md

---
base_model:
- black-forest-labs/FLUX.1-dev
pipeline_tag: text-to-image
language:
- en
license_name: flux-1-dev-non-commercial-license
license: apache-2.0
tags:
- gguf-comfy
---
## startup kit for comfy pack
![screenshot](https://raw.githubusercontent.com/calcuis/gguf-comfy/master/gguf.png)

get the comfy [pack*](https://github.com/calcuis/gguf-comfy/releases):
- decompress the 7z bundle file: Extract All... (it includes everything you need to run a model)
- download [flux1-dev-Q4_0.gguf](https://huggingface.co/city96/FLUX.1-dev-gguf/blob/main/flux1-dev-Q4_0.gguf) (6.32GB); pull it to ./ComfyUI/models/unet
- download [clip_l.safetensors](https://huggingface.co/comfyanonymous/flux_text_encoders/blob/main/clip_l.safetensors) (246MB); pull it to ./ComfyUI/models/clip
- download [t5xxl_fp8_e4m3fn.safetensors](https://huggingface.co/comfyanonymous/flux_text_encoders/blob/main/t5xxl_fp8_e4m3fn.safetensors) (4.89GB); pull it to ./ComfyUI/models/clip
- download [ae.safetensors](https://huggingface.co/black-forest-labs/FLUX.1-schnell/blob/main/ae.safetensors) (319MB); pull it to ./ComfyUI/models/vae
- run the .bat file in the main directory (it will activate the py backend as well as the js frontend)
- drag [gguf-workflow.json](https://github.com/calcuis/gguf-comfy/blob/main/gguf-workflow.json) to the activated browser

*the latest update supports stable diffusion 3.5; please see details [here](https://huggingface.co/calcuis/sd3.5-large-gguf)
#### references: [black-forest-labs](https://github.com/black-forest-labs) [comfyanonymous](https://github.com/comfyanonymous/ComfyUI) [city96](https://github.com/city96/ComfyUI-GGUF)