Update README.md

README.md

@@ -82,7 +82,7 @@ After installing `mistral_inference`, a `mistral-demo` CLI command should be ava
 > ```sh
 > pip install mistral_inference
 > pip install mistral-demo
-> 
+> pip install accelerate #GPU A100/L4
 > pip install git+https://github.com/huggingface/transformers.git
 > ```
 
@@ -90,6 +90,10 @@ If you want to use Hugging Face `transformers` to generate text, you can do some
 
 ```py
 from transformers import AutoModelForCausalLM, AutoTokenizer
+from accelerate import Accelerator #Use only GPU A100/L4
+
+accelerator = Accelerator() #Use only GPU A100/L4
+
 model_id = "EpistemeAI/Fireball-12B"
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(model_id)