Update app.py

app.py

@@ -6,7 +6,7 @@ from huggingface_hub import login
 # Fetch token from environment (automatically loaded from secrets)
 hf_token = os.getenv("gemma3")
 login(hf_token)
-
+# Initialize the client with your model
 client = InferenceClient("hackergeek98/gemma-finetuned")
 
 def respond(
@@ -30,21 +30,20 @@ def respond(
     # Adding the new user message
     messages.append({"role": "user", "content": message})
 
-    # Initialize the response string
-    response = ""
+    # Prepare the prompt for generation
+    prompt = " ".join([msg["content"] for msg in messages])
 
-    # Corrected method for chat completion
-    for message in client.chat_completion(
-        messages=messages,  # Argument should be named 'messages'
+    # Call the Inference API for text generation (or chat completion if supported)
+    response = client.completion(
+        model="hackergeek98/gemma-finetuned",  # Specify the model
+        prompt=prompt,
         max_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-        stream=True  # Stream the response
-    ):
-        # Accumulate the response from the streaming output
-        token = message.choices[0].delta.content
-        response += token
-        yield response
+    )
+
+    # The response will contain the generated text
+    return response["choices"][0]["text"]
 
 # Gradio interface setup
 demo = gr.ChatInterface(