ai
ai copied to clipboard
Continue generation and regeneration
a way to continue generation when the response is over max tokens.
right now a dev can call complete directly to regenerate. but a function for this with a boolean on the request body for is regeneration or not, would be a nice QOL feature
Kinda related to https://github.com/vercel-labs/ai/issues/167#issuecomment-1604790534.
#2164 adds support for requesting a generation without adding another user message while using handleSubmit