Flux ControlNet (Depth, Canny, Hed) - Work 100%

แชร์
ฝัง
  • เผยแพร่เมื่อ 17 ม.ค. 2025

ความคิดเห็น • 37

  • @marcihuppi
    @marcihuppi 5 หลายเดือนก่อน +4

    Error occurred when executing XlabsSampler:
    'ControlNetFlux' object has no attribute 'load_device'
    i already did a git pull to update comfyui... any other ideas?
    thanks in advance ♥

  • @senoharyo
    @senoharyo 5 หลายเดือนก่อน

    Very nice, have tried it last night. Are these controlnets works with flux Checkpoint work flow? :)

  • @dameguy_90
    @dameguy_90 5 หลายเดือนก่อน +4

    Thank you very much for the tutorial, but I am not getting the same quality as yours, just very, very poor quality pictures. And my flux model keeps drawing only animations and not real. Is there a solution?

    • @Huang-uj9rt
      @Huang-uj9rt 5 หลายเดือนก่อน +2

      Yes, I also manipulated it after watching this video to get a terrible image, not as good as the one I got after running flux on mimicpc. I think I'm going to be a big fan of mimicpc from now on, it has all the popular AI tools that I can try for free!

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน +2

      The difference is likely in your settings. Please go to the x-flux-comfyui folder and try the company's pre-built workflows for Canny,Depth, and HED with the default settings

    • @plainpixels
      @plainpixels 5 หลายเดือนก่อน

      Still seems to suck unless you use the same type of images as their examples

    • @adriands8207
      @adriands8207 4 หลายเดือนก่อน

      @@CgTopTips what settings you mean? we are just following all steps and settings in the video but the results are awful

    • @ralf.starke
      @ralf.starke 3 หลายเดือนก่อน

      Same problem here. Only comic paintings, poor quality. No photo realistic pictures.

  • @VaradRane-p2q
    @VaradRane-p2q 5 หลายเดือนก่อน +1

    Can we use this with inpainting techniques ? Is there any workflow for it in ComfyUI ?

  • @CasasYLaPistola
    @CasasYLaPistola 5 หลายเดือนก่อน

    Thanks for the video. One question, does it only work with the Dev model? Does it not work with the schnell model?

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน +1

      Yes, both the Schnell and Dev models work fine

  • @kevinwang7340
    @kevinwang7340 5 หลายเดือนก่อน +1

    the workflow is fine but the result is very off from your samples, it seems like it does not interpret well the inputs and gives you weird images.

  • @shadowheg
    @shadowheg 5 หลายเดือนก่อน +2

    dont working on Dev fp32 with 32gb ram, 4080 12gb VRAM, but working on NF4. but result unsuccessfu (

  • @anagnorisis2024
    @anagnorisis2024 5 หลายเดือนก่อน

    is there a workflow based on this, where i can input an image and do a style or composition transfer?

  • @eltalismandelafe7531
    @eltalismandelafe7531 5 หลายเดือนก่อน

    In the node Canny Edge you have set the resolution to 768. My image is 1280 x 720, how can I set the resolution of the Canny Edge to 1280 x 720 or to get a 1280 x 720 image?

  • @VazgenAkopov1976
    @VazgenAkopov1976 5 หลายเดือนก่อน +2

    It's a pity, but it DOESN'T WORK on 32 gigabytes of RAM and 8 gigabytes of video card memory!(((

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน +3

      Yes, at least 12gb 😕

  • @CGFUN829
    @CGFUN829 5 หลายเดือนก่อน

    Thanks, what resolution you recomend when doing animation using sd1.5 , depth , canny ?

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน +1

      512 or 768

  • @roylow1292
    @roylow1292 5 หลายเดือนก่อน +3

    VAE Docode eror Error occurred when executing VAEDecode:
    Given groups=1, weight of size [4, 4, 1, 1], expected input[1, 16, 128, 128] to have 4 channels, but got 16 channels instead
    File "D:\ComfyUI-aki-v1.3\execution.py", line 316, 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-aki-v1.3\execution.py", line 191, 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-aki-v1.3\execution.py", line 168, in _map_node_over_list
    process_inputs(input_dict, i)
    File "D:\ComfyUI-aki-v1.3\execution.py", line 157, in process_inputs
    results.append(getattr(obj, func)(**inputs))
    File "D:\ComfyUI-aki-v1.3
    odes.py", line 284, in decode
    return (vae.decode(samples["samples"]), )
    File "D:\ComfyUI-aki-v1.3\comfy\sd.py", line 322, in decode
    pixel_samples[x:x+batch_number] = self.process_output(self.first_stage_model.decode(samples).to(self.output_device).float())
    File "D:\ComfyUI-aki-v1.3\comfy\ldm\models\autoencoder.py", line 199, in decode
    dec = self.post_quant_conv(z)
    File "D:\ComfyUI-aki-v1.3\python\lib\site-packages\torch
    n\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
    File "D:\ComfyUI-aki-v1.3\python\lib\site-packages\torch
    n\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
    File "D:\ComfyUI-aki-v1.3\comfy\ops.py", line 93, in forward
    return super().forward(*args, **kwargs)
    File "D:\ComfyUI-aki-v1.3\python\lib\site-packages\torch
    n\modules\conv.py", line 460, in forward
    return self._conv_forward(input, self.weight, self.bias)
    File "D:\ComfyUI-aki-v1.3\python\lib\site-packages\torch
    n\modules\conv.py", line 456, in _conv_forward
    return F.conv2d(input, weight, bias, self.stride,

  • @antoinesaal2372
    @antoinesaal2372 4 หลายเดือนก่อน

    Help ; Error : No module named 'controlnet_aux' (for img2img)
    (My comfyUI, custom_nodes are updated. I watched different controlnet tutos.)
    (I'm using Flux1-dev-Q4_K_S + comfyUI + Controlnet)
    There is a error during the node "Canny edge" : No module named 'controlnet_aux'
    I installed everything that I need (ComfyUImanager, Xlabs, ControlNet auxiliary models, Controlnet Canny) I don't understand why this doesn't work.
    If you have a solution :)

  • @cameochan7405
    @cameochan7405 5 หลายเดือนก่อน

    AttributeError: 'DoubleStreamBlock' object has no attribute 'processor'
    what's this mean pls?

  • @yaahooali
    @yaahooali 5 หลายเดือนก่อน

    Thank you

  • @wowforeal
    @wowforeal 5 หลายเดือนก่อน +1

    Work w NF4?

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน +1

      Yes !

  • @李云-f1b
    @李云-f1b 4 หลายเดือนก่อน

    Very good,

  • @brunocandia9671
    @brunocandia9671 5 หลายเดือนก่อน

    ThX!

  • @joneschunghk
    @joneschunghk 5 หลายเดือนก่อน

    You are installing "requirements.txt" to your python, not python_embeded of comfyui.

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน

      follow x-flux-comfyui installing instruction on githhub page

    • @davoodice
      @davoodice 5 หลายเดือนก่อน

      Yes .your way is not for comfyui portable. ​@@CgTopTips

  • @ismgroov4094
    @ismgroov4094 5 หลายเดือนก่อน

    ❤😅

  • @RiiahTV
    @RiiahTV 5 หลายเดือนก่อน

    its like that!

  • @kallamamran
    @kallamamran 5 หลายเดือนก่อน +1

    Thanks, but... !!! Exception during processing!!! Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument mat1 in method wrapper_CUDA_addmm)

    • @tetsuooshima832
      @tetsuooshima832 5 หลายเดือนก่อน

      I had the exact same error, but today ComfyUI has been updated to support Flux controlnets, so hopefully we don't need this anymore

  • @MilesBellas
    @MilesBellas 5 หลายเดือนก่อน +1

    it works !?!😊

    • @CgTopTips
      @CgTopTips  5 หลายเดือนก่อน +1

      I'm glad you were able to get a result. This workflow need more VRAM to avoid the "Cuda Out of Memory" error

  • @davoodice
    @davoodice 5 หลายเดือนก่อน

    Installation of the package x-flux is not correct. You installed xflux in stand alone python not in comfyUI portable python.