QwQ-32B-preview / app.py
kz919's picture
Update app.py
5485aaf verified
from typing import List, Tuple, Union
from web_ui import WebUI
import math
import os
from qwen_agent.agents import Assistant
from qwen_agent.gui.gradio import gr
def app_gui():
# Define the agent
bot = Assistant(llm={
'model': "QwQ-32B-Preview",
'model_server': 'https://api.sambanova.ai/v1',
'api_key': os.environ.get("SAMBANOVA_API_KEY"),
'generate_cfg': {
'max_input_tokens': 32768,
'max_retries': 10,
'temperature': float(os.environ.get("T", 0.001)),
'repetition_penalty': float(os.environ.get("R", 1.0)),
"top_k": int(os.environ.get("K", 20)),
"top_p": float(os.environ.get("P", 0.8)),
}},
name='QwQ-32B-preview',
description='QwQ-32B-Preview is an experimental research model developed by the Qwen Team, focused on advancing AI reasoning capabilities. As a preview release, it demonstrates promising analytical abilities while having several important limitations such as code switching and recursive reasoning loops. Only single-turn queries are supported in this demo.',
system_message= 'You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step.',
rag_cfg={'max_ref_token': 32768, 'rag_searchers': []},
)
chatbot_config = {
'input.placeholder': "Type \"/clear\" to clear the history",
'verbose': True,
'prompt.suggestions': [
{
'text': 'How many r in strawberry'
},
{
'text': 'Find the least odd prime factor of $2019^8+1$.'
},
{
'text': '''S先生、P先生、Q先生他们知道桌子的抽屉里有16张扑克牌:红桃A、Q、4 黑桃J、8、4、2、7、3 草花K、Q、5、4、6 方块A、5。约翰教授从这16张牌中挑出一张牌来,并把这张牌的点数告诉 P先生,把这张牌的花色告诉Q先生。这时,约翰教授问P先生和Q 先生:你们能从已知的点数或花色中推知这张牌是什么牌吗?于是,S先生听到如下的对话:
P先生:我不知道这张牌。
Q先生:我知道你不知道这张牌。
P先生:现在我知道这张牌了。
Q先生:我也知道了。
请问:这张牌是什么牌?'''
},
]
}
WebUI(bot, chatbot_config=chatbot_config).run(concurrency_limit=80)
if __name__ == '__main__':
app_gui()