Update app.py

app.py

@@ -1,29 +1,40 @@
 import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import AutoTokenizer, AutoConfig, PhiForCausalLM
 import torch
 
-# Load your model from Hugging Face Hub
-model_id = "Anabury/My_finetuned_Phi-4"
+model_id = "Anabury/My_Finetuned_Phi-4"
+
+# Load config to confirm model type
+config = AutoConfig.from_pretrained(model_id)
+
+# Load tokenizer
 tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(
+
+# Use PhiForCausalLM for Phi-4 architecture
+model = PhiForCausalLM.from_pretrained(
     model_id,
+    device_map="auto",
     torch_dtype=torch.float16,
-    device_map="auto"
+    trust_remote_code=True  # if needed for custom implementations
 )
 
+model.config.use_cache = True  # enables faster inference
+
+# Define the chat interface
 def chat(message, history):
-    # encode input
     inputs = tokenizer(message, return_tensors="pt").to(model.device)
-    # generate output
-    outputs = model.generate(**inputs, max_new_tokens=200, pad_token_id=tokenizer.eos_token_id)
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=200,
+        pad_token_id=tokenizer.eos_token_id
+    )
     reply = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    # store history
     history.append((message, reply))
     return history, history
 
 with gr.Blocks() as demo:
     chatbot = gr.Chatbot()
-    msg = gr.Textbox(placeholder="Ask me anything...")
+    msg = gr.Textbox(placeholder="Type your message here...")
     clear = gr.Button("Clear")
 
     msg.submit(chat, [msg, chatbot], [chatbot, chatbot])