Update app.py

app.py

@@ -1,70 +1,83 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import spaces
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
+# --- Model Loading ---
+model_name = "gitglubber/Qwen3-IWM"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
+)
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="gitglubber/Qwen3-IWM")
+# --- System Message ---
+# Define the persona or instructions for the model
+system_message = """"You are an expert at the terminal. Others need you to answer, appropriately. Continue your ownership of the terminal. Output commands decorated with @command. Be verbose in your output. If you need to execute a second command to fulfill the user action @command2. The user will give scenarios."""
 
+# --- Generation Function ---
+@spaces.GPU(duration=120)
+def generate_response(chat_history):
+    # Prepare the model input from the chat history
+    # The system message is the first entry
     messages = [{"role": "system", "content": system_message}]
+    
+    # Add previous user/assistant messages
+    for user_msg, assistant_msg in chat_history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
 
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
+    # Apply the chat template
+    text = tokenizer.apply_chat_template(
         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
+        tokenize=False,
+        add_generation_prompt=True,
+    )
+    model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
 
+    # Generate text
+    generated_ids = model.generate(
+        **model_inputs,
+        max_new_tokens=8192
+    )
+    output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
+    
+    content = tokenizer.decode(output_ids, skip_special_tokens=True)
+    return content
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=4096, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+# --- Gradio Interface ---
+with gr.Blocks(fill_height=True) as demo:
+    gr.Markdown("# IWM Chat Bot")
+    # We use a state object to store the system message, though it's constant here
+    chatbot = gr.Chatbot(scale=1)
+    msg = gr.Textbox(label="Input", scale=0)
+    clear = gr.Button("Clear")
 
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
+    def respond(message, chat_history):
+        if not message.strip(): # Check for empty or whitespace-only messages
+            return "", chat_history
+        
+        # Append the new user message to the history
+        chat_history.append((message, None))
+        
+        # Prepare history for the model (without the last empty spot)
+        model_input_history = chat_history[:-1]
+        model_input_history.append((message, None)) # Add current message for context
+        
+        # Flatten the history for the model function
+        flat_history = []
+        for user, assistant in chat_history:
+            if user: flat_history.append((user, assistant))
+            
+        bot_response = generate_response(flat_history)
+        
+        # Update the last entry in chat_history with the bot's response
+        chat_history[-1] = (message, bot_response)
+        
+        return "", chat_history
 
+    msg.submit(respond, [msg, chatbot], [msg, chatbot])
+    clear.click(lambda: None, None, chatbot, queue=False)
 
-if __name__ == "__main__":
-    demo.launch()
+# Launch the app
+demo.launch()