A new Shortcuts action ‘Use Model’ that lets Shortcuts to run LLM models directly, has three options of models to choose: On-device, Cloud Compute, and ChatGPT model.
There is rate limits for Cloud Compute and ChatGPT models, but there is no clear information for exact rate.
What is the rate limit for Cloud Compute and ChatGPT models?
When does the limit and usage be reset?
iPhone 15 Pro Max
Hi!
Both Cloud Compute and ChatGPT models in Shortcuts do have rate limits to prevent excessive usage, but Apple has not publicly specified exact request or token limits at this time.
Use Apple Intelligence in Shortcuts on iPhone - Apple Support
while general API rate limiting principles apply, Apple has not published explicit numbers or reset timing for 'Use Model' in Shortcuts' Cloud Compute or ChatGPT models. Reset timing may be based on rolling windows, app lifecycle events, or daily limits. For precise details, monitoring usage or contacting Apple support is recommended.
If further detailed rate limits for the ChatGPT model integrated with Shortcuts are needed, checking OpenAI account limits or official Apple Shortcuts do***entation upon updates might provide more clarity
Hi!
Both Cloud Compute and ChatGPT models in Shortcuts do have rate limits to prevent excessive usage, but Apple has not publicly specified exact request or token limits at this time.
Use Apple Intelligence in Shortcuts on iPhone - Apple Support
while general API rate limiting principles apply, Apple has not published explicit numbers or reset timing for 'Use Model' in Shortcuts' Cloud Compute or ChatGPT models. Reset timing may be based on rolling windows, app lifecycle events, or daily limits. For precise details, monitoring usage or contacting Apple support is recommended.
If further detailed rate limits for the ChatGPT model integrated with Shortcuts are needed, checking OpenAI account limits or official Apple Shortcuts do***entation upon updates might provide more clarity
Thanks!
By the way, the ChatGPT model limit in the Shortcuts is separate with OpenAI account limit, it has minutely limit about ~10RPM.
About rate limits of ‘Use Model’ actions in the Shortcuts
