Upload app.py with huggingface_hub

app.py

@@ -1,7 +1,6 @@
 # -*- coding: utf-8 -*-
 import os
 import json
-import re
 import requests
 import gradio as gr
 
@@ -10,7 +9,14 @@ import gradio as gr
 # ----------------------------------------------------------------------
 GROQ_API_KEY = os.environ.get("GROQ_API_KEY")
 GROQ_ENDPOINT = "https://api.groq.com/openai/v1/chat/completions"
-MODEL_NAME = os.environ.get("MODEL_NAME", "llama-3.1-8b-instant")
+
+# Default model can be overridden by the user via the dropdown
+DEFAULT_MODEL = os.environ.get("MODEL_NAME", "llama-3.1-8b-instant")
+AVAILABLE_MODELS = [
+    "llama-3.3-70b-versatile",
+    "llama-3.1-8b-instant",
+    "qwen/qwen3-32b"
+]
 
 # ----------------------------------------------------------------------
 # CSS (will be passed to demo.launch)
@@ -35,7 +41,7 @@ CSS = """
 # ----------------------------------------------------------------------
 # Helper to call Groq API
 # ----------------------------------------------------------------------
-def call_groq(messages):
+def call_groq(messages, model):
     """
     Sends a chat completion request to Groq and returns the assistant's reply.
     """
@@ -46,7 +52,7 @@ def call_groq(messages):
         "Content-Type": "application/json"
     }
     payload = {
-        "model": MODEL_NAME,
+        "model": model,
         "messages": messages,
         "temperature": 0.7
     }
@@ -55,5 +61,5 @@ def call_groq(messages):
     data = response.json()
     # Extract the assistant's content
     content = data["choices"][0]["message"]["content"]
-    # Clean possible markdown code fences
+    # Remove surrounding markdown fences if present
     if content.startswith("