import gradio as gr
from transformers import AutoTokenizer, AutoModel

def chatglm2(q):
    tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True)
    model = AutoModel.from_pretrained("THUDM/chatglm2-6b", trust_remote_code=True, device='cpu')
    model = model.eval()
    response, history = model.chat(tokenizer, q)
    return response
page = gr.Interface(fn=chatglm2, inputs=[gr.Text(label="请提问")],outputs=gr.Label(label="回答："))
page.launch()