Update app.py

app.py

@@ -71,9 +71,8 @@ HF_TOKEN = os.getenv("HF_TOKEN")
 if not HF_TOKEN:
     raise ValueError("HF_TOKEN is missing.")
 
-# Create HF client
+# Initialize client
 client = InferenceClient(
-    provider="nscale",
     api_key=HF_TOKEN
 )
 
@@ -85,7 +84,6 @@ system_message = (
     "based on their requirements."
 )
 
-# Chat function
 def chat_function(message, history):
 
     messages = [
@@ -99,7 +97,6 @@ def chat_function(message, history):
 
     for item in history:
 
-        # Modern Gradio messages format
         if isinstance(item, dict):
 
             role = item.get("role")
@@ -111,7 +108,6 @@ def chat_function(message, history):
                     "content": content
                 })
 
-        # Older tuple fallback
         elif isinstance(item, (list, tuple)) and len(item) == 2:
 
             user_msg, assistant_msg = item
@@ -128,7 +124,6 @@ def chat_function(message, history):
                     "content": assistant_msg
                 })
 
-    # Current user message
     messages.append({
         "role": "user",
         "content": message
@@ -137,7 +132,7 @@ def chat_function(message, history):
     try:
 
         completion = client.chat.completions.create(
-            model="Qwen/Qwen2.5-Coder-7B-Instruct",
+            model="Qwen/Qwen2.5-Coder-7B-Instruct:nscale",
             messages=messages,
             max_tokens=2048,
             temperature=0.7,