fixed response bug

app.py

@@ -4,14 +4,7 @@ from huggingface_hub import InferenceClient
 
 HF_TOKEN = os.getenv("HF_TOKEN")
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
+def respond(message, history, system_message, max_tokens, temperature, top_p):
     client = InferenceClient(model="Nihal2000/gemma3-merged", token=HF_TOKEN)
 
     messages = [{"role": "system", "content": system_message}]
@@ -19,19 +12,25 @@ def respond(
     messages.append({"role": "user", "content": message})
 
     response = ""
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
-        response += token
-        yield response
+    try:
+        for msg in client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            choices = msg.choices
+            token = ""
+            if len(choices) and choices[0].delta.content:
+                token = choices[0].delta.content
+            response += token
+            yield response
+    except Exception as e:
+        yield f"[Error] {str(e)}"
+
+    if not response:
+        yield "[No response from model]"
 
 
 chatbot = gr.ChatInterface(