Update app.py

app.py

@@ -1,14 +1,16 @@
 import os
-import openai
+from openai import OpenAI
 from dotenv import load_dotenv
 
-# Load environment variables (make sure you have a .env file)
+# Load environment variables (e.g., API keys)
 load_dotenv()
 
-# OpenAI Client Initialization using OpenAI's official API
-openai.api_key = os.getenv("TOKEN")
+# Initialize the OpenAI client with Hugging Face API
+client = OpenAI(
+    base_url="https://api-inference.huggingface.co/v1",
+    api_key=os.getenv("TOKEN")
+)
 
-# Model Selection
 model = "meta-llama/Meta-Llama-3-8B-Instruct"
 
 def get_debyez_prompt_template(customer_message):
@@ -18,24 +20,26 @@ def get_debyez_prompt_template(customer_message):
     : '{customer_message}'
     """
 
-# Function to interact with the OpenAI model
+# API or CLI that will interact with the model
 def get_response(prompt):
     try:
         messages = [{"role": "user", "content": prompt}]
-        # Call OpenAI's chat completion API
-        response = openai.ChatCompletion.create(
+        stream = client.chat.completions.create(
             model=model,
             messages=[
                 {"role": m["role"], "content": get_debyez_prompt_template(m["content"])}
                 for m in messages
             ],
             temperature=0.5,
+            stream=True,
             max_tokens=3000,
         )
-        
-        # Extract and return the model's reply
-        return response['choices'][0]['message']['content']
-        
+        response = ""
+        for chunk in stream:
+            response += chunk.choices[0].delta.content or ""
+        if not response.strip():
+            response = "Sorry, I couldn't generate a response right now."
+        return response
     except Exception as e:
         return f"An error occurred: {str(e)}"