Update README.md

README.md

@@ -17,7 +17,8 @@ pipeline_tag: image-to-image
 library_name: diffusers
 ---
 
-> **Note:** This is a repackaged version of the [black-forest-labs/FLUX.2-dev](https://huggingface.co/black-forest-labs/FLUX.2-dev) model. The `flux2-dev.safetensors` file, which was part of the same model found in the `transformer/` folder, was removed because loading the model caused two transformers and OMM errors, even though the model theoretically fit on the GPU. The file size decreased from 178GB with the duplicated load to around 110GB when performing an inference.
+> **Note:** This is a repackaging of the [black-forest-labs/FLUX.2-dev](https://huggingface.co/black-forest-labs/FLUX.2-dev) model. Only the `flux2-dev.safetensors` file located in the root directory was removed, as it contained the same model as the one defined in the `transformer/` folder. Because of this, `diffusers` was loading the transformer twice, causing out-of-memory (OOM) errors. By removing this file, the duplicate loading was avoided and memory usage during inference was reduced from approximately **178 GB** to **110 GB**, enabling stable execution.
+
 
 ![Teaser](./teaser_generation.png)
 ![Teaser](./teaser_editing.png)