FLEX

U have three options

1. Fp16 (15.21 GB)

2. Q8 (8.08 GB)

3. Q6_K(6.24 GB)

(my experience the result i got from in-painting here /model/1108146/fluxfill-devgguf

and controlnet here

/model/1488208/flux1-dev-controlnet-union-pro-20fp8

more professional but this model have good features because its very flexible)

THis has Free license So please Support who spend time and train this model for u

(Ostris)

the original model https://huggingface.co/ostris/Flex.2-preview

u can find this at end of model page

---

i prepare professional WORKFLOW one here

/model/1514962?modelVersionId=1713872

U can drop any image for workflow

Dont forget 👍

Flux Loras works

(Redux not work)

---

Diffusion model

• Place the Flex.2-preview model in: ComfyUI/models/diffusion_models/

Text encoders

• Place the following in ComfyUI/models/text_encoders/:

  • clip_l.safetensors

  • Option 1 (FP8): t5xxl_fp8_e4m3fn_scaled.safetensors

  • Option 2 (FP16): t5xxl_fp16.safetensors

VAE

• Place ae.safetensors in: ComfyUI/models/vae/

Required Custom Node

• Install ComfyUI-FlexTools:

cd ComfyUI/custom_nodes
git clone https://github.com/ostris/ComfyUI-FlexTools

ComfyUI/

├── models/

│ ├── diffusion_models/

│ │ └─── flex.2-preview.safetensors

│ ├── text_encoders/

│ │ ├─── t5xxl_fp8_e4m3fn_scaled.safetensors

│ │ ├─── t5xxl_fp16.safetensors

│ │ └─── clip_l.safetensors

│ └── vae/

│ └── ae.safetensors

├── custom_nodes/

└── ComfyUI-FlexTools/ (git clone from repository)