Update app.py

app.py

@@ -1,28 +1,19 @@
 import gradio as gr
-import torch
-from transformers import AutoTokenizer
-from generate import load_model, generate_text
+from generate.py import generate_response
 
-# Settings
-MODEL_PATH = "evo_decoder.pt"
-DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
-TOKENIZER_NAME = "gpt2"  # Or your custom tokenizer
-MAX_TOKENS = 50
+def chat_with_evo(user_input):
+    response = generate_response(user_input)
+    return response
 
-# Load tokenizer & model
-tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_NAME)
-vocab_size = tokenizer.vocab_size
-model = load_model(vocab_size, MODEL_PATH, DEVICE)
+with gr.Blocks() as demo:
+    gr.Markdown("## 🤖 EvoDecoder Chatbot (Lightweight Conversational AI)")
+    with gr.Row():
+        with gr.Column(scale=4):
+            user_input = gr.Textbox(label="Ask Evo anything...", placeholder="Type your question here...")
+            submit_btn = gr.Button("Submit")
+        with gr.Column(scale=6):
+            response_output = gr.Textbox(label="Evo's Response")
 
-def chat(prompt):
-    output = generate_text(model, tokenizer, prompt, max_new_tokens=MAX_TOKENS, device=DEVICE)
-    return output
+    submit_btn.click(fn=chat_with_evo, inputs=user_input, outputs=response_output)
 
-# Gradio UI
-iface = gr.Interface(fn=chat,
-                     inputs=gr.Textbox(label="Ask Evo..."),
-                     outputs=gr.Textbox(label="Evo's Response"),
-                     title="🧠 EvoDecoder Chatbot",
-                     description="A lightweight conversational model powered by EvoDecoder")
-
-iface.launch()
+demo.launch()