Update README.md

README.md

@@ -9,12 +9,8 @@ tags:
 
 # Boogu-Image-0.1-Turbo GGUF
 
-This repository contains Flat Quantized GGUF versions of the newly released **Boogu-Image-0.1-Turbo** foundation model. 
 
-These models have been successfully packaged with the `omnigen` architecture tag, allowing them to bypass `llama-quantize` limitations and load directly into ComfyUI via the `ComfyUI-GGUF` node suite. 
-
-### Required Text Encoder & VAE
-To run the Boogu architecture properly, you cannot use standard SD or Flux encoders. You must download the specific VAE and multimodal text encoders. 
+### Required Text Encoder & VAE 
 
 * **Text Encoder (Qwen3-VL):** You must use an FP8 scaled version of the Qwen3-VL encoder. *(Note: Do NOT use the NF4 versions floating around—NF4 strips out the visual `mmproj` tower, which will instantly crash your workflow with a shape mismatch).* 
   * Download the FP8 encoder from the [Comfy-Org Qwen3-VL Repo](https://huggingface.co/Comfy-Org/Qwen3-VL).