irmagon's picture
Update README.md
a621b5b verified
---
license: mit
---
12.02.2026
- update Z-Image-Turbo presets with Qwen3-4b-Z-Image-Turbo-AbliteratedV1 clip encoder
Download and unpack into user_presets from https://huggingface.co/irmagon/ComfyFooocusPlus_1.0.9/resolve/main/zit_v3.zip?download=true
22.01.2026
- ADD FLUX2-Klein-4B support
Need to manual download models from https://civitai.com/api/download/models/2600878?type=Model&format=SafeTensor&size=pruned&fp=fp8
21.01.2026
- ADD FLUX2-Klein-9B support
1. Backup FooocusPlusAI and python_embedded folders!!!
2. Rename \FooocusPlusAI\comfy folder to comfy1
3. Download https://huggingface.co/irmagon/ComfyFooocusPlus_1.0.9/resolve/main/FooocusPlusAI_F2K.7z
4. Unpack it into you FooocusPlus folders with replace (notice: \FooocusPlusAI\backend_base\params_mapper.py will be replaced for work change resolution)
5. Run requirements install with run_Comfy_Install.bat
6. Set correct path in \FooocusPlusAI\comfy\extra_model_paths.yaml
7. Run with run_FooocusPlus_without_update.bat
8. Select presets FLUX2-Klein
Used models:
https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-fp8/resolve/main/flux-2-klein-9b-fp8.safetensors?download=true
https://huggingface.co/Comfy-Org/vae-text-encorder-for-flux-klein-9b/resolve/main/split_files/text_encoders/qwen_3_8b_fp8mixed.safetensors?download=true
https://huggingface.co/Comfy-Org/vae-text-encorder-for-flux-klein-9b/resolve/main/split_files/vae/flux2-vae.safetensors?download=true
31.12.2025
- ADD UltraFLUX VAE
- ADD comfyui-cache-dit nodes (https://github.com/xlite-dev/comfyui-cache-dit)
wthout DIT - [00:07<00:00, 1.08it/s]![image](https://cdn-uploads.huggingface.co/production/uploads/65244a2378ed2f9ac1344e07/vTg4Y0vkMB05l92K87rM4.png)
with DIT - [00:03<00:00, 2.53it/s]![image](https://cdn-uploads.huggingface.co/production/uploads/65244a2378ed2f9ac1344e07/4lVeWIQC_ndaO6pr76nzI.png)
1. Download https://huggingface.co/irmagon/ComfyFooocusPlus_1.0.9/resolve/main/FooocusPlusAI_DIT.7z
2. Unpack it into you FooocusPlus folders with replace files
3. Select presets ZI-TurboDIT, ZI-Turbo_AIODIT
25.12.2025 (long way)
1. Backup FooocusPlusAI and python_embedded folders!!!
2. Rename \FooocusPlusAI\comfy folder to comfy0
3. Download https://huggingface.co/irmagon/ComfyFooocusPlus_1.0.9/resolve/main/FooocusPlusAI.7z
4. Unpack it into you FooocusPlus folders
5. Run requirements install with run_Comfy_Install.bat
6. Set correct path in \FooocusPlusAI\comfy\extra_model_paths.yaml
7. Run with run_FooocusPlus_without_update.bat
If "not found module" error in terminal window - install it with oX-COMMAND.bat
You can run internal ComfyUI with run_Comfy_gpu.bat (edit or run directly any workflow, save changes with Export API file)
Select presets in ZITurbo category or Favorites
![image](https://cdn-uploads.huggingface.co/production/uploads/65244a2378ed2f9ac1344e07/6Z44B28l26xTMsn1G3v3j.png)