Hey!
My current projects involves the use of lots of client chat completion requests to gpt-4o-audio-preview with audio output.
Not only is this method pretty expensive, the token limit is also hindering progress on the project. The completion time does not really matter so I dont care about the 24h wait.
From the docs alone I am having trouble understanding if it is possible to use the batch API for this usecase.
Any help is appreciated!
1 Like