Flux Outpainting in ComfyUI

Поделиться
HTML-код
  • Опубликовано: 19 дек 2024

Комментарии • 28

  • @romanioamd5319
    @romanioamd5319 2 месяца назад +1

    I confirm everything works fine. Simple workflow and easy to understand the best of the tested others did not like it, because got of poor results. Thanks.

  • @michael-c4t
    @michael-c4t 2 месяца назад

    Instant sub for having good audio/files readily available/interface easily legible in the video.

  • @TestMyHomeChannel
    @TestMyHomeChannel 2 месяца назад +1

    it is incredible! Definitely, you are a Master! Thank you for sharing the workflow!

    • @AIFuzz59
      @AIFuzz59  2 месяца назад

      @@TestMyHomeChannel anytime! You are a Master!!

  • @realarnoldpalmer
    @realarnoldpalmer 2 месяца назад

    Excellent! This workflow is amazing! Thank you

  • @pto2k
    @pto2k 2 месяца назад +1

    Thanks for sharing.
    Wish your next house be a happier house.

  • @LabTeamStudios
    @LabTeamStudios 2 месяца назад

    gracias por crear estos tutoriales

  • @adsfcsdsd
    @adsfcsdsd Месяц назад

    That AI voice is really great if it is one :D

  • @stevietee3878
    @stevietee3878 2 месяца назад +1

    Hi, thank you for this, however when I use the workflow (using the same models) I receive the following error: " KSampler
    mat1 and mat2 shapes cannot be multiplied (7000x16 and 64x3072)". I've updated Comfy and tried different input image sizes and different models, without success, do you have a fix please ?

    • @AIFuzz59
      @AIFuzz59  2 месяца назад +1

      @@stevietee3878 that error usually
      Means that the vae or Lora and model do not match. Make sure you are using the correct ones for Flux

  • @MilesBellas
    @MilesBellas 2 месяца назад +1

    Outpainting = incredibly powerful
    .
    Style Transfer = final hurdle ?
    "Open Flux" was just released: an improved version of Schnell ?

  • @Rokyadome
    @Rokyadome 2 месяца назад +1

    Thank You!

  • @hotlineoperator
    @hotlineoperator 2 месяца назад +2

    My favorite host voice in RUclips

    • @AIFuzz59
      @AIFuzz59  2 месяца назад

      @@hotlineoperator awwwww ❤️❤️

    • @goodie2shoes
      @goodie2shoes 2 месяца назад

      His name is Rupert and he lives in his moms basement

    • @hotlineoperator
      @hotlineoperator 2 месяца назад

      @@goodie2shoes But I like the voice what Rupert uses.

    • @goodie2shoes
      @goodie2shoes 2 месяца назад

      @@hotlineoperator I can respect that!

  • @giuseppedaizzole7025
    @giuseppedaizzole7025 Месяц назад

    Hi there, there is something i dont understand, u made a Flux outpainting, but the workflow is with sdxl model and also i can see something like, set sdxl image. and controlnet union promax is a sdxl model. where is flux here, can u explain me please?

  • @mehdi6166
    @mehdi6166 2 месяца назад

    my flux model is at unet folder and not showing in the workflow
    Any solution please?

    • @AIFuzz59
      @AIFuzz59  2 месяца назад

      @@mehdi6166 place it in checkpoint folder

    • @mehdi6166
      @mehdi6166 2 месяца назад

      @@AIFuzz59 I did but it didn't work

    • @NicholasCarn
      @NicholasCarn 2 месяца назад

      you have the wrong flux model, there's a more recent one which goes in the checkpoint folder, same thing happened to me

    • @mehdi6166
      @mehdi6166 2 месяца назад

      @@NicholasCarn thanks , what is the huggingface id?

  • @zinheader6882
    @zinheader6882 2 месяца назад

    what is the name of the voice model you are using?.. 😂 rvc?

  • @shuntera
    @shuntera 2 месяца назад +1

    Doesnt work at all!

    • @AIFuzz59
      @AIFuzz59  2 месяца назад +1

      @@shuntera where are you having issues?

    • @romanioamd5319
      @romanioamd5319 2 месяца назад +1

      I confirm everything works fine i tested.

    • @shuntera
      @shuntera 2 месяца назад

      @@AIFuzz59 I got this error:
      no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
      !!! Exception during processing !!! Only zero batch or zero channel inputs are supported, but got input shape: [1, 1, 0, 0]
      Traceback (most recent call last):
      File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 323, in execute
      output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 198, in get_output_data
      return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 169, in _map_node_over_list
      process_inputs(input_dict, i)
      File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 158, in process_inputs
      results.append(getattr(obj, func)(**inputs))
      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "D:\ComfyUI_windows_portable\ComfyUI
      odes.py", line 356, in encode
      mask_erosion = torch.clamp(torch.nn.functional.conv2d(mask.round(), kernel_tensor, padding=padding), 0, 1)
      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      RuntimeError: Only zero batch or zero channel inputs are supported, but got input shape: [1, 1, 0, 0]