Update Wan Trained Loras README with file details

Loras/Wan Trained Loras/README.md

@@ -4,4 +4,33 @@ This folder contains LoRA models trained with WAN (Weighted Attention Network).
 
 ## Files
 
-Upload your WAN high and low LoRA files here.
+- **Bia_training_high_noise.safetensors** - WAN trained LoRA with high noise (293 MB)
+- **Bia_training_low_noise.safetensors** - WAN trained LoRA with low noise (293 MB)
+
+## Usage
+
+### High Noise vs Low Noise
+
+- **High Noise**: Better for more creative/varied outputs with higher guidance
+- **Low Noise**: Better for more controlled/precise outputs with lower guidance
+
+### ComfyUI
+
+1. Download the desired LoRA file (high or low noise)
+2. Place it in your ComfyUI `models/loras/` folder
+3. Use the "Load LoRA" node in your workflow
+4. Recommended strength: 0.8-1.0
+
+### Automatic1111/Forge
+
+1. Download the desired LoRA file
+2. Place it in your `models/Lora/` folder
+3. Use in prompt: `<lora:Bia_training_high_noise:1.0>` or `<lora:Bia_training_low_noise:1.0>`
+4. Adjust weight as needed (0.6-1.2 range)
+
+## Training Info
+
+- **Training Date**: October 4, 2025
+- **Model Type**: WAN (Weighted Attention Network)
+- **Base Model**: Stable Diffusion
+- **File Format**: SafeTensors