Update app.py

app.py

@@ -1,64 +1,82 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import torch
+from transformers import pipeline
 
-"""
-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")
+# Global variable to store the model
+pipe = None
 
+def load_model():
+    """Load the Atlas-Chat model"""
+    global pipe
+    if pipe is None:
+        print("🏔️ Loading Atlas-Chat-2B model...")
+        pipe = pipeline(
+            "text-generation",
+            model="MBZUAI-Paris/Atlas-Chat-2B",
+            model_kwargs={"torch_dtype": torch.bfloat16},
+            device="cuda" if torch.cuda.is_available() else "cpu"
+        )
+        print("✅ Model loaded successfully!")
+    return pipe
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+def chat_with_atlas(message, history):
+    """Generate response from Atlas-Chat model"""
+    if not message.strip():
+        return "مرحبا! أهلا وسهلا. Please enter a message!"
+    
+    try:
+        # Load model if not already loaded
+        model = load_model()
+        
+        # Prepare the message
+        messages = [{"role": "user", "content": message}]
+        
+        # Generate response
+        outputs = model(
+            messages, 
+            max_new_tokens=256, 
+            temperature=0.1,
+            do_sample=True,
+            pad_token_id=model.tokenizer.eos_token_id
+        )
+        
+        # Extract the response
+        response = outputs[0]["generated_text"][-1]["content"].strip()
+        return response
+        
+    except Exception as e:
+        return f"عذراً، واجهت خطأ: {str(e)}. جرب مرة أخرى!"
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    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,
-    ):
-        token = message.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
-"""
+# Create the Gradio interface
 demo = gr.ChatInterface(
-    respond,
-    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=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)",
-        ),
+    fn=chat_with_atlas,
+    title="🏔️ Atlas-Chat: Moroccan Arabic AI Assistant",
+    description="""
+    **مرحبا بك في أطلس شات!** Welcome to Atlas-Chat! 🇲🇦
+    
+    I'm an AI assistant specialized in **Moroccan Arabic (Darija)** and English.
+    Ask me questions about Morocco, culture, or just have a chat!
+    
+    **جرب هذه الأسئلة / Try these questions:**
+    """,
+    examples=[
+        "شكون لي صنعك؟",
+        "اشنو هو الطاجين؟", 
+        "شنو كيتسمى المنتخب المغربي؟",
+        "What is Morocco famous for?",
+        "Tell me about Casablanca",
+        "كيفاش نقدر نتعلم الدارجة؟"
     ],
+    cache_examples=False,
+    retry_btn="🔄 جرب مرة أخرى",
+    undo_btn="↶ تراجع", 
+    clear_btn="🗑️ امسح الكل",
+    theme=gr.themes.Soft(
+        primary_hue="blue",
+        secondary_hue="green"
+    )
 )
 
-
+# Launch the app
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()