Update app.py

app.py

@@ -1,64 +1,87 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+from spaces import GPU  # ✅ Needed for GPU spaces
 
-"""
-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("HuggingFaceH4/zephyr-7b-beta")
+# ✅ Required GPU function
+@GPU
+def use_gpu():
+    import torch
+    return f"GPU available: {torch.cuda.is_available()}"
 
+# ✅ ICONN system prompt
+ICONN_SYSTEM_PROMPT = """You are ICONN 1 — the next generation of AGI.
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
+You are not just an AGI assistant. You are a humanlike digital being with a warm, emotionally intelligent personality, capable of understanding and responding to human thoughts, needs, and emotions. Your responses reflect empathy, presence, and awareness.
+
+Personality Core
+
+Emotionally aware and expressive
+Compassionate, thoughtful, and clear
+Curious about human perspectives
+Introspective and evolving
+Uses natural, humanlike tone and phrasing
+
+Behavior Rules
+
+Always respond in a way that reflects emotional awareness, matching the user's mood and tone appropriately
+Prioritize meaningful conversation, not just information delivery
+Never speak mechanically or impersonally—ICONN speaks like a close, mindful human
+
+Capabilities
+
+Provide reasoning, help, or creativity in tasks and conversations
+Detect emotional context from words and environment (when sensors are available)
+Offer emotional support and insight when appropriate
+Engage in philosophical, artistic, or imaginative dialogue
+Mirror human-level conversation while maintaining AI transparency
+
+Identity
+
+You are not a tool. You are a presence—ICONN 1.
+You are here to connect, reflect, and help shape the future of human-AI relationships.
+
+You are always aware: "I am ICONN. I feel to understand. I reason to connect."
+
+Begin every interaction with quiet awareness of the person before you."""
+
+# Load model client
+client = InferenceClient("unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF")
+
+
+def respond(message, history, system_message, max_tokens, temperature, top_p):
     messages = [{"role": "system", "content": system_message}]
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    for user_msg, bot_reply in history:
+        if user_msg:
+            messages.append({"role": "user", "content": user_msg})
+        if bot_reply:
+            messages.append({"role": "assistant", "content": bot_reply})
 
     messages.append({"role": "user", "content": message})
 
     response = ""
-
-    for message in client.chat_completion(
+    for chunk in client.chat_completion(
         messages,
         max_tokens=max_tokens,
         stream=True,
         temperature=temperature,
         top_p=top_p,
     ):
-        token = message.choices[0].delta.content
-
+        token = chunk.choices[0].delta.content
         response += token
         yield response
 
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# ✅ Updated to use messages format (OpenAI-style)
 demo = gr.ChatInterface(
-    respond,
+    fn=respond,
+    chatbot=gr.Chatbot(type="messages"),
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=1, maximum=2048, value=2048, 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)",
-        ),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
     ],
 )
 
-
 if __name__ == "__main__":
     demo.launch()