Instructions to use oumoumad/LumiPic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use oumoumad/LumiPic with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Qwen/Qwen-Image-Edit-2511,black-forest-labs/FLUX.2-klein-base-4B,black-forest-labs/FLUX.2-klein-base-9B", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("oumoumad/LumiPic") prompt = "Turn this cat into a dog" input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png") image = pipe(image=input_image, prompt=prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
Error while running
I encountered another error while I was trying your workflow "SDR_to_HDR_Klein9b":
"RuntimeError: ERROR: Could not detect model type of: D:\ComfyUI_windows_portable\ComfyUI\models\unet\A2R_Klein_Standard.safetensors" but this model IS present in the Unet-folder!
How can this be fixed? What is meant by the term "model type"? Thanks in advance!
Is that definitely the model? Could it be Lora?
Is that definitely the model? Could it be Lora?
It seems me that you are right, slowly but I start to understand! But now I have another problem. I tried to replicate this workflow from this video: https://www.youtube.com/watch?v=z0ue28hbMTk
In this video all neccessary staeps are described in order to run the workflow but I am not able to change the "Image Edit (Flux2Klein 4B). I downloaded the offical workflow by oumoumad and put it into my ComfyUI but now I cannot change "unet_name", "clip_name" and "vae_name" as decribed from minute 2 on, they all stuck to undesired name despite what model and/or I choose from the drop-down menu. What might be my misktake? Any help I greatly appreciated! Thx
Is that definitely the model? Could it be Lora?
It seems me that you are right, slowly but I start to understand! But now I have another problem. I tried to replicate this workflow from this video: https://www.youtube.com/watch?v=z0ue28hbMTk
In this video all neccessary staeps are described in order to run the workflow but I am not able to change the "Image Edit (Flux2Klein 4B). I downloaded the offical workflow by oumoumad and put it into my ComfyUI but now I cannot change "unet_name", "clip_name" and "vae_name" as decribed from minute 2 on, they all stuck to undesired name despite what model and/or I choose from the drop-down menu. What might be my misktake? Any help I greatly appreciated! Thx
Once you have downloaded the models, and sorted them into the appropriate folders, you need to restart the comfy or press R .
I did all the steps you have suggested but it seems me that the "SDR to HDR" node seems to be "locked" I looked for an "unlock" menu entry but did succeed! How to unlock the node? Thx
I did all the steps you have suggested but it seems me that the "SDR to HDR" node seems to be "locked" I looked for an "unlock" menu entry but did succeed! How to unlock the node? Thx
If the nodes are red, you need to install https://github.com/oumad/ComfyUI_Gear
Thank you very much for your patience and your help. Now I am able to run the workflow: I did a complete new installation of ComfyUI and followed all your steps! Perfectly results!!