Update app.py

app.py

@@ -1,44 +1,24 @@
 import gradio as gr
 import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM
-import spaces  # ← обязательно для ZeroGPU
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+import spaces  # обязательно для ZeroGPU
 
-model_name = "openai-community/gpt2"
+model_name = "google/flan-t5-base"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
-model.to("cuda" if torch.cuda.is_available() else "cpu")
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
 
-@spaces.GPU  # 💥 Без этого ZeroGPU не сработает!
+@spaces.GPU
 def respond(message, history=[]):
-    history = history or []
-    full_prompt = "Ты — ассистент, который знает всё об Университете Иннополис.\n"
-
-    for user, bot in history:
-        full_prompt += f"Пользователь: {user}\nБот: {bot}\n"
-    full_prompt += f"Пользователь: {message}\nБот:"
-
-    inputs = tokenizer(full_prompt, return_tensors="pt").to(model.device)
-    outputs = model.generate(
-        **inputs,
-        max_new_tokens=200,
-        do_sample=True,
-        temperature=0.7,
-        top_p=0.9,
-        pad_token_id=tokenizer.eos_token_id
-    )
-    output_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    response = output_text.split("Бот:")[-1].strip()
-
+    prompt = f"Answer the question: {message}"
+    inputs = tokenizer(prompt, return_tensors="pt").to(device)
+    outputs = model.generate(**inputs, max_new_tokens=100)
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
     history.append((message, response))
     return history
 
-chat = gr.ChatInterface(
-    fn=respond,
-    title="Innopolis Bot",
-    chatbot=gr.Chatbot(label="Unitrip"),
-    examples=["Когда основан Университет Иннополис?", "Сколько программ бакалавриата?"],
-    cache_examples=False
-)
+demo = gr.ChatInterface(fn=respond)
 
 if __name__ == "__main__":
-    chat.launch()
+    demo.launch()