Update README.md

README.md

@@ -10,6 +10,8 @@ library_name: diffusers
 Inference includes noise correction based on current timestep, intentional blurring plus upscale interpolation: all with intention of providing as fast-as-possible reconstruction that is viable and consistent with any noise levels.  
 Intended use-case is live-preview during generative model inference.  
 
+Shapes/Channels/etc are inferred from the base VAE, so no configuration changes are needed between different models.
+
 - Model definition and training code [here](https://github.com/vladmandic/sdnext/blob/dev/modules/vae/sd_vae_micro_train.py)
 - Example inference code [here](https://github.com/vladmandic/sdnext/blob/dev/modules/vae/sd_vae_micro.py)
 
@@ -17,14 +19,19 @@ Example using **MicroDecoder** with `Flux.2-Klein-9B` and compared with official
 
 ![MicroDecoder](https://huggingface.co/vladmandic/MicroDecoder/resolve/main/MicroDecoder.jpg)
 
-## Examples
+## Example
 
 ```shell
-sd_vae_micro_train.py --dim 256 --epochs 350 --resolution 512 --scale 8 --lr 0.0003 --folder ~/generative/Input/vae/ --vae AutoencoderKLQwenImage21 --repo Qwen/Qwen-Image-2.1 --output AutoencoderKLQwenImage21.safetensors
-sd_vae_micro_train.py --dim 256 --epochs 350 --resolution 512 --scale 4 --lr 0.0003 --folder ~/generative/Input/vae/ --vae AutoencoderKLFlux2 --repo black-forest-labs/FLUX.2-klein-9B --output AutoencoderKLFlux2.safetensors
-sd_vae_micro_train.py --dim 256 --epochs 350 --resolution 512 --scale 4 --lr 0.0003 --folder ~/generative/Input/vae/ --vae AutoencoderKL --repo stabilityai/stable-diffusion-xl-base-1.0 --output AutoencoderKL.safetensors
-sd_vae_micro_train.py --dim 256 --epochs 350 --resolution 512 --scale 4 --lr 0.0003 --folder ~/generative/Input/vae/ --vae AutoencoderKLWan --repo Wan-AI/Wan2.2-I2V-A14B-Diffusers --output AutoencoderKLWan.safetensors
-sd_vae_micro_train.py --dim 256 --epochs 350 --resolution 512 --scale 4 --lr 0.0003 --folder ~/generative/Input/vae/ --vae AutoencoderKLQwenImage --repo krea/Krea-2-Turbo --output AutoencoderKLQwenImage.safetensors
+sd_vae_micro_train.py \
+  --dim 256 \
+  --epochs 350 \
+  --resolution 512 \
+  --scale 4 \
+  --lr 0.0003
+  --folder ~/generative/Input/vae/ \
+  --vae AutoencoderKLQwenImage21 \
+  --repo Qwen/Qwen-Image-2.1
+  --output MicroVAE-qwen21.safetensors
 ```
 
 ```log
@@ -41,5 +48,5 @@ Epoch  | Train Tot | Tr PSNR | Tr SSIM | Val Tot  | Val PSNR | Val SSIM | Val L1
 350/350  | 2.3010    | 28.51   | 0.8511  | 0.9781   | 35.83    | 0.9734   | 0.0111  | 0.0278  *
 
 Training complete! Best validation PSNR: 35.83 dB (Epoch 350)
-Saved best EMA model weights to: AutoencoderKLFlux2.safetensors
+Saved best EMA model weights to: MicroVAE-qwen21.safetensors
 ```