Stefan GussnerNew blog post about how we're using the Streams API to autofill forms as data is being parsed by an llm
Turning JSON Objects from a Fetch Response into an Async Generator with the Streams API | text-to-json Blog
When interacting with large language models, we often want to present the user with content before the full response is available. We can use the Streams API to parse partial responses on the fly and display them to the user.