I have a problem with getting the results from API Assistant GPT. I suppose that there is some bug in the OpenAI side, but that’s a regular situation for them.
The point is that Assistant is answering with 2 cutted messages today (I checked it in the playground), but in the “Result” in the Integration it shows only the second small part, so it doesn’t takes the first part of the message.
How to fix it? We need to have in the “Result” both parts of the Assistant Answer/
I have the same problem, which started about 1 hour ago. I checked in the playground, and in the thread, the message is sometimes split into two or three parts. However, the response from the API only includes the last part of the message, and the previous parts are missing, making it impossible to retrieve them.
I’m experiencing this as well today!
I expect to get a JSON response but it cuts up the JSON into several messages making it impossible to parse without changing to code to start going through all messages and try to build the entire response…
Sometimes it just responds that it seems to have technical issues.
I can add that, from my perspective, this violates the very idea of the assistant, which is designed to eliminate the need to write a system for managing context and messages, as everything is supposed to happen under the hood.
Same problem here. Code blocks are broken, model 0125 falls back in laziness and creates dummy or pseudocode and says, I can insert my logic (as Role: Software Developer to create full scripts without truncation)
(days before, he was able to create a full python script).
I’m not sure, it is adjustable. But this error occurs again and again over a month. I’m not sure, is there somebody min the team, which modified it to save his job as dev 
And to ask every and every time again: read complete scripts, with n=15000 and many more commands: the stupid guy read only again and again only the first rows or signs.
And again and again: GPT must remind itself: “ups, i must load python or something else”, error, error, error. Repeating rounds again and again for the same instructions. It’s like a child, where the parents repeat things 10 times. 
Same thing happening for me too, Suddenly JSON output is with 2 cutted messages
Im having the same issue, have been getting partial responses intermittently since yesterday evening but has been much worse today
I am having exact same problem!
Same problem here since yesterday evening… !
Is everyone still facing the same issue? 
I solved the problem by processing and merging several messages from the list into one, and everything works as before.
Yeah, I’m still having the same problem
I am still having the issue, So does OpenAI Dev team aware of this…?
Yeah that works, but as you said before its so awkward and not intuitive. I’d rather just drop assistants and use the chat API.
I am not sure if they are aware of this bug. I just tested and the Assistant works fine on the playground, but its broken trough API calls…
I hope they check this forum at least once a day?!
Just got the same question why certain responses only seemed to start at the second part.
this on top of assistant file connect issues (4-0125)
Sorry for the troubles, we’re looking into this now.
I’m experiencing this as well, my thread_id: thread_iW7jjCR8yQ8VjSlU6UeM8cKY
