Edit images locally with an open model
Use the official FLUX.2 Klein 4B ComfyUI image-editing template to transform an input image while keeping source files on your machine.
Before you begin
- Difficulty
- Intermediate
- Software
- ComfyUI
- Hardware
- Use the same 4B text encoder, diffusion model, and VAE family as the official template. Memory depends on input size and model variant.
Sources and files
Official ComfyUI edit templates and required filesChoose a model for this task
The steps below use the recommended model. Alternatives have their own package and command; open their model pages before switching.
The workflow
Install and update ComfyUI
Install ComfyUI locally and update it before importing the FLUX.2 Klein 4B editing template.
Load the official editing template
Choose FLUX.2 Klein 4B Image Edit in ComfyUI Workflow Templates. The guide has both base and fast distilled variants; start with the distilled variant for a shorter first run.
Download the template’s files
For the distilled 4B template get qwen_3_4b.safetensors, flux-2-klein-4b-fp8.safetensors, and flux2-vae.safetensors from the official guide.
Place each model file
Put qwen_3_4b.safetensors in models/text_encoders, flux-2-klein-4b-fp8.safetensors in models/diffusion_models, and flux2-vae.safetensors in models/vae under your ComfyUI directory.
Load an image and describe the edit
Load an input image into the template’s LoadImage node, then write a focused change such as changing a material or removing one object. Confirm the selected diffusion file matches the distilled template.
Run and compare
Queue the workflow and compare the saved output to the input. Confirm the requested change is visible and unrelated parts remain useful; adjust the prompt or reference images when needed.
When it doesn’t go to plan
Restart ComfyUI after placing files and compare each exact filename and folder with the official FLUX.2 Klein guide.
Check the LoadImage node and its links in the official edit template, then make the prompt describe the desired change to the supplied image.
The model behind this workflow
FLUX.2 Klein · 4BWill it run on your machine?
Save your machine to see a personalized rating and its reasoning.
Add my hardwareKeep exploring
Set up a private AI chat assistant
Install a local runtime, run Qwen3.5 9B, confirm responses, and know when to choose the smaller 4B package.
Build a private local coding assistant
Connect an open coding-capable model in Ollama to Cline, run a small repository task, and review the result locally.
Generate images on your own GPU
Use the official FLUX.2 Klein 4B ComfyUI template with exact model files, a first prompt, and an output check.