J S
J S
@pebeto Would you please consider merging the PR to have a working minor version release? The bug introduced in 0.4.5 (the ajax in path) still persists.
+1! We already have HTTP.jl 1.10 and JSON 0.21
Out of curiousity, what is the use case here? What would be the advantages of implementing it? It seems that it basically parses JSON twice anyway, once inside a try-catch...
> Thank you for the insight, @svilupp! I'm thinking of a "default" streamcallback function for word-by-word streaming that would be very useful for a user for a number of reasons....
Perhaps we should learn from what others are doing? Langchain: > astream_events > > It’s worth calling out that we’re using our new astream_eventsmethod to easily stream back all events...
Btw I’ll tackle gpt-4 vision first! Also, I forgot to add the finetuning API - I’ll add that to the list.
A few updates from my end, the functionality I've tried so far does not require any changes to the `create_chat` (unless we want to add nicer interfaces). # Vision comprehension...
I’ve commented on the above mentioned thread. I personally know very little about the streaming use case - everything I do doesn’t need it. I’m familiar with it only for...
I do agree that we should create a separate issue for adding some examples (incl. for the new OpenAI functionality), but perhaps this issue can be closed now?
At the moment, we have 14 test cases across 2 categories. The ideal number would be at least 5 categories with at least 5 test cases each.