I have a pretty low level openai subscription and an constantly plagued by rate limiting.
I simply tell ai to continue and it does but is there a way to effectively slow down the rate of AI requests so the rate limit threshold is not exceeded
Mate, sounds like OpenAI’s giving you the “slow down, tiger” treatment, you’re basically trying to feed a Mini Cooper the workload of a freight train.
You can throttle requests in Wappler, but honestly, with your Ambassador badge, I’m surprised OpenAI hasn’t just handed you unlimited tokens out of respect.
But seriously, we all run into this. Whether you’re using AI inside Wappler or outside it, the trick is to slow the calls down a bit, so you don’t smash the rate‑limit wall.
I tried to tell wappler to generate a medium size cms, after about 20 rate limit messages followed by me telling it to continue it was clear it mas not even close to finished and was goingbto take days.
Can wappler really be sending out 2000+ RPM?
Hi Brian,
Have you tried OpenAI's Codex? You can connect to the project folders on your local PC. This allows Codex to review and update the code in Wappler project directly.
A ChatGPT Plus subscription will give you access to their latest models and I have found Plus subscription to be quite generous in token limits and usage when using the Sol-* models.
Strangely i am not getting the rate limit error today. Also still not sold of the code produced (tried chatGPT and Deepseek to date)
Thats what i have currently
Thanks for the reply @guptast
I normally use all the AI models outside the Wappler application but decided to try the integrated AI with wappler. Still have reservations on how effective it is.
Results have been questionable.
