Currently the API has a Create Thread & Run, which is great for quickly starting to use an assistant, but to continue using said assistant it takes 2 calls: Create Message and Create Run
It would be awesome to combine these, and would prevent the extra loading time between awaiting the Create Message response
Hi and welcome to the Developer Forum!
That seems like a solid idea, cheers for the suggestion.
Was just looking this up, nowhere in the documentation yet. I’m sure its coming soon as the API is still in the beta phase.
this is a good idea hopefully the amount of tokens go down this way cause making a thread and then making a message and running it takes me 3000 tokens kind of expensive
Glad to see this feedback on here. Because we can only upload user messages to the thread, essentially every use case would require following up add message with a run
Waiting for this update, needed to reduce 2 api calls to 1.
How is this not a thing yet?