ComfyUI: FLUX + ControlNet + IPAdapter (SDXL Version) Integration

Поділитися
Вставка
  • Опубліковано 25 січ 2025

КОМЕНТАРІ • 21

  • @India.stories
    @India.stories 2 місяці тому

    Thanks so much! with your tutorial I was finally able to understand style transfer using ip adpater

  • @Huang-uj9rt
    @Huang-uj9rt 5 місяців тому

    For a beginner also say that your videos are really very friendly, thank you very much. Because of my professional needs and the high learning threshold of flux, I've been using mimicpc to run flux before, it can load the workflow directly, I just want to download the flux model, and it handles the details wonderfully, but after watching your video, I'm using mimicpc to run flux again finally have a different experience, it's like I'm starting to get started! I feel like I'm starting to get the hang of it.

  • @RKKrish-p1b
    @RKKrish-p1b 5 місяців тому +2

    awesome tutorial thanks ...

  • @Rachelcenter1
    @Rachelcenter1 5 місяців тому +1

    5:55 but why do you have positive/negative prompt in the middle of the workflow and then another prompt box on the far left hand side of the workflow?

    • @AINxtGen8
      @AINxtGen8  4 місяці тому

      Thank you for asking.
      The prompt on the left is for creating a realistic, black and white style photo, while the prompt on the right side of the workflow is for instructing how to create an illustration-style image. I could reuse the prompt on the left side and connect it directly, but this would cause conflicting results between the instruction prompt and the IPAdapter, which follows the illustration style. You can use the same prompt, but in such a case, you should remove the specific description of the photo style and only describe the object itself. This approach will help avoid conflicts and ensure consistency in the final output.

  • @eveekiviblog7361
    @eveekiviblog7361 4 місяці тому

    you can also try with cpds controlnet and add lora for flux. You can use Florence 2 and Olama LLMs to remove style from first text bar. BTW I really expect to see comparison of GGUF (lets say Q8) and NF4 flux models comparison using Union!

  • @fabioroncal8763
    @fabioroncal8763 3 місяці тому +1

    it just doesnt work... i can get those Krita and the checkpoint. models..it would be more helful if you put the links to all the necessary assets

  • @Kirmm
    @Kirmm 3 місяці тому

    I'm stuck on Anyline Preprocessor. Just can't import it. Says (IMPORT FAILED). Tried "Try Fix" button, tried Uninstalling and reinstalling but no go. Any tips?
    EDIT: OKAY fixed it with AIO Aux Preprocessor.
    But now I get an error with ClipVision.
    I downloaded the IP-Adapter models into a IPAdapter folder but still no go.
    Error occurred when executing IPAdapterUnifiedLoader:
    ClipVision model not found.
    File "C:\COMFY-UI\ComfyUI\execution.py", line 152, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "C:\COMFY-UI\ComfyUI\execution.py", line 82, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "C:\COMFY-UI\ComfyUI\execution.py", line 75, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "C:\COMFY-UI\ComfyUI\custom_nodes\ComfyUI_IPAdapter_plus-main\IPAdapterPlus.py", line 559, in load_models
    raise Exception("ClipVision model not found.")

  • @Rachelcenter1
    @Rachelcenter1 5 місяців тому

    when i open the workflow it says "set node,. get node and anyLineProcessor" are missing, so i go to "install missing" and it prompts me to install "anyline", program restarts, i hit refresh on browser and its still missing.

    • @AINxtGen8
      @AINxtGen8  4 місяці тому

      Alternatively, you can use the "AIO Aux Preprocessor" node from the "ControlNet Auxiliary Preprocessors" extension (it has similar functionality).
      imgur.com/qBdKLcH
      github.com/Fannovel16/comfyui_controlnet_aux

  • @filterophilicxx5914
    @filterophilicxx5914 5 місяців тому

    which llm did you use to convert image to prompt? is it open source please guide

    • @AINxtGen8
      @AINxtGen8  5 місяців тому +2

      I use monica.im (it is a multi-AI integration service) like POE. You can refer to using opensource LLM in the following few nodes, all have similar functions:
      github.com/gokayfem/ComfyUI_VLM_nodes
      github.com/kijai/ComfyUI-Florence2
      github.com/pythongossss/ComfyUI-WD14-Tagger
      github.com/miaoshouai/ComfyUI-Miaoshouai-Tagger

    • @SebAnt
      @SebAnt 5 місяців тому

      Excellent tutorial!!

  • @HimanshuChaudhari-y9t
    @HimanshuChaudhari-y9t 5 місяців тому +3

    Krita diffusion tutorial

  • @BrowneNorton-h5u
    @BrowneNorton-h5u 4 місяці тому

    Thomas Ronald Williams Brenda Lopez Susan

  • @kronosiii9379
    @kronosiii9379 5 місяців тому +1

    Thank you for the tutorial. AnylinePreprocessor by themisto does not load even after update all comfyui and a manual installation , pinokio\api\comfyui.git\app\custom_nodes\ComfyUI-Anyline

    • @AINxtGen8
      @AINxtGen8  5 місяців тому

      After installation, try restarting Pinokio to see if it resolves the issue. Since I don't use Pinokio, I'm not sure about the exact cause. Alternatively, you can use the "AIO Aux Preprocessor" node from the "ControlNet Auxiliary Preprocessors" extension (it has similar functionality).
      imgur.com/qBdKLcH

    • @kronosiii9379
      @kronosiii9379 5 місяців тому +1

      @@AINxtGen8 Thank you for the quick reply! I'll verify this soon!