API tools faq
paste
Guest User

V2 Manual Comfyui Flux Models Install

a guest
Nov 2nd, 2024
14,235
2
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 5.83 KB | None | 2 0
raw download clone embed print report
  1. Manual Install Method from my last Youtube Video: https://youtu.be/zLGthxUe2jc
  2.  
  3.  
  4. LOCAL INSTALL:
  5. Download Comfyui portable version here:https://github.com/comfyanonymous/ComfyUI
  6.  
  7. Inside the ComfyUI_windows_portable\ComfyUI\custom_nodes:
  8. git clone https://github.com/ltdrdata/ComfyUI-Manager.git
  9.  
  10. Inside the models/unet folder, download one of the GGUF model:
  11. https://huggingface.co/city96/FLUX.1-dev-gguf/tree/main
  12.  
  13. Inside the models/clip folder, download one of the GGUF text encoder model:
  14. https://huggingface.co/city96/t5-v1_1-xxl-encoder-gguf/tree/main
  15.  
  16. or/and one of the original Text encoders: https://huggingface.co/comfyanonymous/flux_text_encoders/tree/main
  17.  
  18. As well as the brand new ViT-L-14-TEXT-detail-improved-hiT-GmP-TE-only-HF.safetensors text encoder:
  19. https://huggingface.co/zer0int/CLIP-GmP-ViT-L-14
  20.  
  21. Inside the models/upscale_models folder, download the upscaler model:
  22. https://openmodeldb.info/models/4x-ClearRealityV1
  23.  
  24. Inside the models/xlabs/controlnet folder (if it doesn't exist yet you can just create it), download the 3 V3 controlnet models:
  25. https://huggingface.co/XLabs-AI/flux-controlnet-collections/tree/main
  26.  
  27. Inside the models/vae folder, download the vae model:
  28. https://huggingface.co/Aitrepreneur/FLX/resolve/main/ae.safetensors
  29.  
  30.  
  31.  
  32.  
  33. IN ADDITION FOR THE VERSION 2 OF THE WORKFLOW WE NEED A BIT MORE STUFF...
  34.  
  35. REACTOR/INSIGHTFACE INSTALLATION:
  36. One of the very first things we need to do once comfyui is extracted and the extraction is verified, we need to go to the ComfyUI_windows_portable\python_embeded folder and do a few things:
  37.  
  38. First, inside that folder is a python.exe file, if you run it, it should give you the python version used for this virtual python env, it's usually something like python 11 but you never know, when I run the python.exe file, I get this message:
  39. "Python 3.11.8 (tags/v3.11.8:db85d51, Feb 6 2024, 22:03:32) [MSC v.1937 64 bit (AMD64)] on win32
  40. Type "help", "copyright", "credits" or "license" for more information.
  41. >>> "
  42. where you can see that this python version is 3.11.8, so from there, we need the script to be able to read the python version correctly and depending on the version that is written it needs to download the appropriate file inside the python_embeded folder.
  43.  
  44. If the python version if 3.9, this file needs to be downloaded inside the python_embeded folder: https://huggingface.co/Aitrepreneur/FLX/resolve/main/insightface-0.7.3-cp39-cp39-win_amd64.whl?download=true
  45.  
  46. If the python version if 3.10, this file needs to be downloaded inside the python_embeded folder:
  47. https://huggingface.co/Aitrepreneur/FLX/resolve/main/insightface-0.7.3-cp310-cp310-win_amd64.whl?download=true
  48.  
  49. If the python version if 3.11, this file needs to be downloaded inside the python_embeded folder:
  50. https://huggingface.co/Aitrepreneur/FLX/resolve/main/insightface-0.7.3-cp311-cp311-win_amd64.whl?download=true
  51.  
  52. If the python version if 3.12, this file needs to be downloaded inside the python_embeded folder:
  53. https://huggingface.co/Aitrepreneur/FLX/resolve/main/insightface-0.7.3-cp312-cp312-win_amd64.whl?download=true
  54.  
  55.  
  56.  
  57. Then once this is done, you need to run the following commands:
  58.  
  59. python.exe -m pip install --use-pep517 facexlib
  60.  
  61.  
  62. python.exe -m pip install git+https://github.com/rodjjo/filterpy.git
  63.  
  64.  
  65. python.exe -m pip install onnxruntime==1.19.2 onnxruntime-gpu==1.15.1 insightface-0.7.3-cp311-cp311-win_amd64.whl
  66.  
  67.  
  68. don't forget to replace here "insightface-0.7.3-cp311-cp311-win_amd64.whl" by the correct file name version that you donwloaded
  69.  
  70. then continue with the rest of the models.
  71.  
  72.  
  73.  
  74.  
  75.  
  76. Also, next,
  77.  
  78. go inside the ComfyUI_windows_portable\ComfyUI\custom_nodes and:
  79.  
  80. git clone https://github.com/balazik/ComfyUI-PuLID-Flux.git
  81.  
  82.  
  83.  
  84. then go to the ComfyUI_windows_portable\ComfyUI\models
  85.  
  86. and create a folder called "pulid" and download inside this file:
  87. https://huggingface.co/Aitrepreneur/FLX/resolve/main/pulid_flux_v0.9.0.safetensors?download=true
  88.  
  89.  
  90. also inside the models folder git clone this repo from huggingface:
  91. git lfs install
  92. git clone https://huggingface.co/Aitrepreneur/insightface
  93.  
  94. this should create a "insightface" folder with a bunch of models inside
  95.  
  96. then clone this repo from huggingface:
  97. git lfs install
  98. git clone https://huggingface.co/Aitrepreneur/CogVideo
  99.  
  100. then cd into the CogVideo\CogVideoX-5b-I2V\transformer folder and dl those 3 files inside:
  101. https://huggingface.co/Aitrepreneur/FLX/resolve/main/diffusion_pytorch_model-00001-of-00003.safetensors?download=true
  102. https://huggingface.co/Aitrepreneur/FLX/resolve/main/diffusion_pytorch_model-00002-of-00003.safetensors?download=true
  103. https://huggingface.co/Aitrepreneur/FLX/resolve/main/diffusion_pytorch_model-00003-of-00003.safetensors?download=true
  104.  
  105. also inside the models/clip folder, download these files inside:
  106. https://huggingface.co/Aitrepreneur/FLX/resolve/main/longclip-L.pt?download=true
  107.  
  108. https://huggingface.co/Aitrepreneur/FLX/resolve/main/t5xxl_fp8_e4m3fn.safetensors?download=true
  109.  
  110.  
  111. then in checkpoints dl this model:
  112. https://huggingface.co/Aitrepreneur/FLX/resolve/main/sd_xl_base_1.0_0.9vae.safetensors?download=true
  113.  
  114.  
  115. then in controlnet dl this model:
  116.  
  117. https://huggingface.co/Aitrepreneur/FLX/resolve/main/diffusion_pytorch_model_promax.safetensors?download=true
  118.  
  119.  
  120.  
  121.  
  122.  
  123. RUNPOD INSTALL:
  124.  
  125. Create a new account: https://bit.ly/runpodAi
  126.  
  127. Choose A 24gb Vram GPU and choose the aitrepreneur/comfyui template.
  128. Click on the port 8888 and do the same thing as with the local install and place all the correct models into the correct folder using the following command:
  129.  
  130. curl -L -o “full name of the file” “URL”
  131. Ex: curl -L -o "flux1-schnell.sft" "https://huggingface.co/black-forest-labs/FLUX.1-schnell/resolve/main/flux1-schnell.sft?download=true"
  132.  
  133. Also use the command: tail -f comfyui.log
  134. inside the workspace/logs to get the real time logs
Tags: Workflow Flux ComfyUI image generation
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!