Lora-Training-in-Comfy

Lora-Training-in-Comfy requires a minimum of 256MB (0.3GB) of VRAM for base operation. For optimal performance, a GPU with at least 4GB of VRAM is recommended. This node supports low VRAM mode for resource-constrained setups.

✅ RTX 3060 (12GB): Yes, fully compatible with 10.6GB headroom. ✅ RTX 4070 (12GB): Yes, fully compatible with 10.6GB headroom. ✅ RTX 4070 Ti (16GB): Yes, fully compatible with 14.2GB headroom. ✅ RTX 4090 (24GB): Yes, fully compatible with 21.4GB headroom

Lora-Training-in-Comfy requires the following PyTorch-related packages: lion-pytorch, open-clip-torch, pytorch-lightning. Ensure your ComfyUI environment has these installed. Ensure your PyTorch installation matches your CUDA version (use torch.version.cuda to check).

To run Lora-Training-in-Comfy, you need to install: accelerate==0.23.0, altair, bitsandbytes, dadaptation, diffusers, easygui, einops, fastapi, ftfy, gradio==3.41.2, httpx, huggingface-hub, lion-pytorch, lycoris-lora, numpy, open-clip-torch, opencv-python, pandas, pillow, prodigyopt, pytorch-lightning, requests, rich, safetensors, scipy, tensorboard, toml, transformers==4.30.2, uvicorn, voluptuous, wandb, xformers. You can install these using pip or add them to your requirements.txt file.

Lora-Training-in-Comfy supports low VRAM mode. To reduce memory usage: (1) Enable --lowvram or --medvram flags in ComfyUI, (2) Reduce batch size to 1, (3) Use fp16 or fp8 precision if supported, (4) Close other GPU applications.

To install Lora-Training-in-Comfy: (1) Navigate to your ComfyUI/custom_nodes directory, (2) Clone the repository: git clone https://github.com/LarryJane491/Lora-Training-in-Comfy, (3) Install dependencies: pip install -r requirements.txt (if present), (4) Restart ComfyUI. Alternatively, use ComfyUI Manager for one-click installation.