Hi, I have tried the official demo to optimize my prompt based on the Qwen model using TextGrad, but encountered some issues. Could you please check if there is any problem?
environment:
textgrad==0.1.8
openai == 2.14.0
code:
from openai import OpenAI
from textgrad.engine.local_model_openai_api import ChatExternalClient
client = OpenAI(api_key=*****',
base_url='http://lanz.hikvision.com/v3/openai/model')
engine = ChatExternalClient(client=client, model_string="Qwen3-Coder-480B")
print(engine.generate(max_tokens=40, content='who are you ?'))
error:
sqlite3.OperationalError: no such column: "size" - should this be a string literal in single-quotes?