update commit with phi-3 mini 5

app.py

@@ -1,14 +1,19 @@
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+import torch
+
+# Force device based on availability
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+print(f"Using device: {device}")
 
 # Load Phi-3 Mini model
 model_id = "microsoft/phi-3-mini-4k-instruct"
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(
-    model_id, torch_dtype="auto", device_map="auto"
-)
+    model_id, torch_dtype=torch.float16 if device.type == "cuda" else torch.float32
+).to(device)
 
-pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, device=0 if device.type == "cuda" else -1)
 
 # OpenAI-style messages (new format)
 def chat_fn(message, history):
@@ -29,11 +34,10 @@ def chat_fn(message, history):
 
     return reply
 
+# Gradio UI
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     gr.Markdown("## 💬 Chat with Phi-3 Mini")
-    gr.Markdown(
-        "Welcome to your AI Assistant powered by Phi-3 Mini. Ask me anything or request code examples!"
-    )
+    gr.Markdown("Lightweight AI Assistant powered by Microsoft's Phi-3 Mini. Works best with short prompts. Ask away!")
 
     gr.ChatInterface(
         fn=chat_fn,
@@ -43,7 +47,8 @@ with gr.Blocks(theme=gr.themes.Soft()) as demo:
             "Write a JavaScript function to reverse a string.",
             "Explain how transformers work.",
         ],
-        chatbot=gr.Chatbot(type="messages")  # fixes the deprecated tuples warning
+        chatbot=gr.Chatbot(type="messages")
     )
 
-demo.launch(share=True)
+# Launch without SSR and share (for Spaces)
+demo.launch(debug=True, ssr_mode=False)