In this meeting, the State Changers discuss how to use Xano to implement a model in a workflow. Justin explains that he uses API calls to retrieve information from Xano and passes it into the prompt. He adds additional text to the prompt for the desired analysis. The response is then obtained through Xano with the API call and can be analyzed for accuracy. Anthony mentions that working with lists can be challenging, so he creates the response in JSON format and decodes it for structured output. They also discuss using variable style prompts for substitution and tokenization. Overall, they find these methods helpful for using Xano effectively. Additionally, the meeting briefly touches on a personal question asked to Chatty about famous people with similar voices.
(Source: Office Hours Extra - OpenAI 2/10 )
