Add a note about xformers in README

Author: kn

examples/dreambooth/README.md

@@ -317,4 +317,7 @@ python train_dreambooth_flax.py \
   --max_train_steps=800
 ```
 
-You can also use Dreambooth to train the specialized in-painting model. See [the script in the research folder for details](https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint).
+### Training with xformers:
+You can enable memory efficient attention by [installing xFormers](https://github.com/facebookresearch/xformers#installing-xformers) and padding the `--enable_xformers_memory_efficient_attention` argument to the script. This is not available with the Flax/JAX implementation.
+
+You can also use Dreambooth to train the specialized in-painting model. See [the script in the research folder for details](https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint).

examples/text_to_image/README.md

@@ -160,3 +160,6 @@ python train_text_to_image_flax.py \
   --max_grad_norm=1 \
   --output_dir="sd-pokemon-model"
 ```
+
+### Training with xformers:
+You can enable memory efficient attention by [installing xFormers](https://github.com/facebookresearch/xformers#installing-xformers) and padding the `--enable_xformers_memory_efficient_attention` argument to the script. This is not available with the Flax/JAX implementation.

examples/textual_inversion/README.md

@@ -124,3 +124,6 @@ python textual_inversion_flax.py \
   --output_dir="textual_inversion_cat"
 ```
 It should be at least 70% faster than the PyTorch script with the same configuration.
+
+### Training with xformers:
+You can enable memory efficient attention by [installing xFormers](https://github.com/facebookresearch/xformers#installing-xformers) and padding the `--enable_xformers_memory_efficient_attention` argument to the script. This is not available with the Flax/JAX implementation.