Update app.py

app.py

@@ -1,28 +1,25 @@
 import torch
-import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model_path = "cody82/unitrip"  # путь к локальной модели
+model_id = "cody82/unitrip"  # или другой ID модели с Huggingface
 
-config = AutoConfig.from_pretrained(model_path, local_files_only=True)
-tokenizer = AutoTokenizer.from_pretrained(model_path, local_files_only=True)
-model = AutoModelForCausalLM.from_pretrained(
-    model_path,
-    config=config,
-    local_files_only=False,
-    torch_dtype=torch.float32,
-    device_map="auto" if torch.cuda.is_available() else None,
-)
+tokenizer = AutoTokenizer.from_pretrained(model_id)  # без local_files_only
+model = AutoModelForCausalLM.from_pretrained(model_id)  # без local_files_only
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+model.to(device)
 
 system_message = "Ты — умный помощник по Университету Иннополис."
 
-def respond(message, history=None):
-    if history is None:
-        history = []
+while True:
+    user_input = input("🧑 Вопрос: ").strip()
+    if user_input.lower() in ["exit", "выход", "quit"]:
+        break
+
+    prompt = f"{system_message}\nUser: {user_input}\nAssistant:"
 
-    prompt = f"{system_message}\nUser: {message}\nAssistant:"
+    inputs = tokenizer(prompt, return_tensors="pt").to(device)
 
-    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
     with torch.no_grad():
         outputs = model.generate(
             **inputs,
@@ -33,11 +30,6 @@ def respond(message, history=None):
             use_cache=True,
         )
 
-    generated_tokens = outputs[0][inputs["input_ids"].shape[1]:]
-    answer = tokenizer.decode(generated_tokens, skip_special_tokens=True).strip()
-
-    history.append((message, answer))
-    return history
-
-chat = gr.ChatInterface(fn=respond, title="Innopolis Assistant")
-chat.launch()
+    generated = outputs[0][inputs["input_ids"].shape[1]:]
+    answer = tokenizer.decode(generated, skip_special_tokens=True)
+    print(f"🤖 Ответ: {answer.strip()}\n")