Update app.py

app.py

@@ -1,20 +1,35 @@
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM, TextStreamer
 import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+model_name = "microsoft/Phi-4-mini-instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float32)
+
+def chat(user_input, history=[]):
+    messages = [{"role": "system", "content": "Ты полезный ассистент."}]
+    for i, (q, a) in enumerate(history):
+        messages.append({"role": "user", "content": q})
+        messages.append({"role": "assistant", "content": a})
+    messages.append({"role": "user", "content": user_input})
+    
+    inputs = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
+    outputs = model.generate(inputs, max_new_tokens=256, do_sample=True, temperature=0.7)
+    reply = tokenizer.decode(outputs[0], skip_special_tokens=True).split("assistant")[-1].strip()
+    
+    history.append((user_input, reply))
+    return reply, history
 
-model_id = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float32)
-model.to("cpu")
-streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
+with gr.Blocks() as demo:
+    gr.Markdown("# 🤖 Флэри чат (на Phi-4-mini)")
+    chatbot = gr.Chatbot()
+    msg = gr.Textbox(label="Ваш вопрос")
+    state = gr.State([])
 
-def chat_fn(prompt):
-    messages = [{"role": "user", "content": prompt}]
-    input_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-    inputs = tokenizer(input_text, return_tensors="pt").to("cpu")
-    output = model.generate(**inputs, max_new_tokens=200, do_sample=True, temperature=0.7)
-    response = tokenizer.decode(output[0], skip_special_tokens=True)
-    return response.split("user")[-1].strip()
+    def respond(message, history):
+        reply, history = chat(message, history)
+        return chatbot.update(value=history), history
 
-gr.Interface(fn=chat_fn, inputs="text", outputs="text", title="💬 FlareGPT на TinyLlama").launch()
+    msg.submit(respond, [msg, state], [chatbot, state])
 
+demo.launch()