chore: Update flash-attn dependency in app.py and requirements.txt

app.py

@@ -6,8 +6,8 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStream
 import torch
 import gradio as gr
 from threading import Thread
-# import subprocess
-# subprocess.run('pip install -U flash-attn transformers sentencepiece', shell=True)
+import subprocess
+subprocess.run('pip install -U flash-attn', shell=True)
 
 models_available = [
     "MohamedRashad/Arabic-Orpo-Llama-3-8B-Instruct",

requirements.txt

@@ -2,5 +2,4 @@ transformers==4.44.1
 torch
 accelerate==0.33.0
 sentencepiece==0.2.0
-flash-attn==2.6.3
 spaces