V3 Manual Comfyui Flux Models Install

Manual Install Method from my last Youtube Video: https://youtu.be/q5kpr84uyzc


1. Download and extract the ComfyUI portable version from:
   https://github.com/comfyanonymous/ComfyUI

2. After extraction, you should have this folder structure:
   ```
   ComfyUI_windows_portable/
   ├── ComfyUI/
   │   ├── models/
   │   └── custom_nodes/
   ├── python_embeded/
   └── run_nvidia_gpu.bat
   ```

## Python Environment Setup (Important First Step!)

Navigate to `ComfyUI_windows_portable/python_embeded` and:

1. Run `python.exe` to check your Python version
2. Based on your Python version, download the appropriate InsightFace wheel file:
   - Python 3.9: `insightface-0.7.3-cp39-cp39-win_amd64.whl`
   - Python 3.10: `insightface-0.7.3-cp310-cp310-win_amd64.whl`
   - Python 3.11: `insightface-0.7.3-cp311-cp311-win_amd64.whl`
   - Python 3.12: `insightface-0.7.3-cp312-cp312-win_amd64.whl`

   Download from: https://huggingface.co/Aitrepreneur/FLX/tree/main

3. Install required packages:

   python.exe -m pip install --use-pep517 facexlib
   python.exe -m pip install git+https://github.com/rodjjo/filterpy.git
   python.exe -m pip install onnxruntime onnxruntime-gpu insightface-0.7.3-cp311-cp311-win_amd64.whl


don't forget to replace here "insightface-0.7.3-cp311-cp311-win_amd64.whl" by the correct file name version that you donwloaded


## Custom Nodes Installation

Navigate to `ComfyUI_windows_portable/ComfyUI/custom_nodes` and clone these repositories:


git clone https://github.com/ltdrdata/ComfyUI-Manager.git
git clone https://github.com/ssitu/ComfyUI_UltimateSDUpscale --recursive
git clone https://github.com/kijai/ComfyUI-LivePortraitKJ
git clone https://github.com/PowerHouseMan/ComfyUI-AdvancedLivePortrait


## Models Installation

Navigate to `ComfyUI_windows_portable/ComfyUI/models` and create the following structure:

### 1. UNET Models
Create `models/unet` and download:
- Choose one GGUF model from: https://huggingface.co/city96/FLUX.1-dev-gguf/tree/main
- FP8 Models (optional for 24GB+ VRAM):
  - `flux1-dev-fp8.safetensors`
  - `flux1-fill-dev_fp8.safetensors`
  - `flux1-canny-dev-fp8.safetensors`
  - `flux1-depth-dev-fp8.safetensors`

### 2. CLIP Models
Create `models/clip` and download:
- Choose one GGUF encoder from: https://huggingface.co/city96/t5-v1_1-xxl-encoder-gguf/tree/main
- Optional text encoders from: https://huggingface.co/comfyanonymous/flux_text_encoders/tree/main
- `ViT-L-14-TEXT-detail-improved-hiT-GmP-TE-only-HF.safetensors`
- `longclip-L.pt`
- `clip_l.safetensors`
- `llava_llama3_fp8_scaled.safetensors`
- `t5xxl_fp8_e4m3fn.safetensors`
- `t5xxl_fp16.safetensors`

Also clone:

git clone https://huggingface.co/Aitrepreneur/clip-vit-large-patch14

Download `model.safetensors` inside the cloned directory.

### 3. VAE Models
Create `models/vae` and download:
- `ae.safetensors`
- `ae.sft`
- `LTX_vae.safetensors`
- `hunyuan_video_vae_bf16.safetensors`

### 4. XLabs Controlnet Models
Create `models/xlabs/controlnets` and download:
- `flux-canny-controlnet-v3.safetensors`
- `flux-depth-controlnet-v3.safetensors`
- `flux-hed-controlnet-v3.safetensors`

### 5. Controlnet Models
Create `models/controlnet` and download:
- `diffusion_pytorch_model_promax.safetensors`
- `Shakker-LabsFLUX1-dev-ControlNet-Union-Pro.safetensors`

### 6. Upscale Models
Create `models/upscale_models` and download:
- `4x-ClearRealityV1.pth`
- `RealESRGAN_x4plus_anime_6B.pth`

### 7. PuLID Model
Create `models/pulid` and download:
- `pulid_flux_v0.9.0.safetensors`

### 8. LLM Models
Create `models/LLM` and:
```bash
git clone https://huggingface.co/Aitrepreneur/Florence-2-base
git clone https://huggingface.co/Aitrepreneur/Florence-2-large
git clone https://huggingface.co/Aitrepreneur/llava-llama-3-8b-text-encoder-tokenizer
```

Inside `llava-llama-3-8b-text-encoder-tokenizer`, download:
- `model-00001-of-00004.safetensors`
- `model-00002-of-00004.safetensors`
- `model-00003-of-00004.safetensors`
- `model-00004-of-00004.safetensors`

### 9. Style Models
Create `models/style_models` and download:
- `flux1-redux-dev.safetensors`

### 10. Loras
Create `models/loras` and download:
- `csetiarcane-nfjinx-v1-6000.safetensors`

### 11. LivePortrait
Create `models/liveportrait` and download:
- `landmark.onnx`

### 12. SAMs
Create `models/sams` and download:
- `sam_vit_b_01ec64.pth`

### 13. Ultralytics
Create `models/ultralytics` and:

git clone https://huggingface.co/Aitrepreneur/bbox
git clone https://huggingface.co/Aitrepreneur/segm

Download:
- `face_yolov8n.pt`

### 14. Additional Repositories
In the `models` directory:

git lfs install
git clone https://huggingface.co/Aitrepreneur/insightface
git clone https://huggingface.co/Aitrepreneur/CogVideo


Inside `CogVideo/CogVideoX-5b-I2V/transformer`, download:
- `diffusion_pytorch_model-00001-of-00003.safetensors`
- `diffusion_pytorch_model-00002-of-00003.safetensors`
- `diffusion_pytorch_model-00003-of-00003.safetensors`




## RunPod Installation

1. Create account at: https://bit.ly/runpodAi
2. Choose a 24GB VRAM GPU with the aitrepreneur/comfyui template
3. Access port 8888
4. Follow the same folder structure as above
5. Use `curl` to download files:

   curl -L -o "filename" "URL"

6. Monitor logs with:

   tail -f workspace/logs/comfyui.log