It’s been great working with plugins, but one thing isn’t so great from the user experience - the request/response speed.
It seems like the more JSON you provide in both the request/response, the longer it takes ChatGPT to stream in each word. You can see it streaming through and with something like the OpenTable plugin it takes ages.
- Wondering if anyone has any tricks/tips to help speed up the request/response time.
- Also I’m curious why the JSON on both sides is shown to users. I understand showing to developers, but it seems to be taking more time to generate this text as JSON compared to showing the response outright. Maybe an OpenAI fix could be to only use this during developer settings.