1 min+ for assistant API to answer

[info1158]

So the assistant API has been extremely slow the last couple of days. Our chatbots is taking sometimes up to 1 minute to generate a short answer.

My usage tier 2. Our prompt is optimized for efficiency.

Will this issue be fixed?

[merefield]

20, 30 sec assistants API answer

There’s always the option of converting your project to Completions. My bot uses it’s own chain of thought loop and uses only Completions. Answers using smaller models even involving functions are almost always sub 5 seconds or less. And humans probably don’t need a response in less than one second For example, see this 2 second response:

[Screamadelica]

I’ve met the same issue and I described them in my post

It takes too long for API to respond recently

I find Open AI API works really slow on my project, so I wrote the following little demo to test the time it takes for OpenAI to respond: import openai import time import os class Chatbot: def __init__(self): self.question = "Say hello to me." self.client = openai.OpenAI( api_key=os.environ.get("OPENAI_API_KEY"), base_url=os.environ.get("OPENAI_BASE_URL") ) def run(self): while True: try: start…

[merefield]

we really don’t need another topic about this.

@Screamadelica perhaps you could link your post.

[Screamadelica]

I’ve tried searching many keywords but didn’t find this post sry😂I think its replies can probably help and will try them out tomorrow, thanks dude.

[merefield]

I’m referring to this Topic. Thanks for using search !

[Screamadelica]

I’ve tried to do so but it pops out “An error occurred: Sorry, you can’t include links in your posts.”. I’m new to this community and haven’t found out how to fix it yet.
(Update) After reading many posts I’m given the ability to link the post now

[info1158]

Even though this topic has been discussed a lot recently I cant find any solutions for this issue. For our use case we cant use “Chat completions”.

If you have heard of any solutions to this, please let me know.

[niklasohrn139]

yeah having some severe issues with this lately. I SINCERELY hope there will be some update pushed soon, this is not sustainable whatsoever. Lots of people complaining about this.

[merefield]

That’s really silly, ugh. It should definitely allow local links.

[PaulBellow]

@merefield

It should definitely allow local links.

I think it’s because they were a new user.

@niklasohrn139

SINCERELY hope there will be some update pushed soon

The mod team will bring this up at the next meeting if we haven’t heard anything by then.

Thanks for reporting!

[merefield]

@PaulBellow

I think it’s because they were a new user.

Indeed, but still, it’s a little silly not to allow hyperlinks to local content?

[info1158]

@PaulBellow

The mod team will bring this up at the next meeting if we haven’t heard anything by then.

Thanks for reporting!

Alright, since we have some clients waiting for the chatbots to get live. Do you have a timeline for when we can expect some clarity in this matter? When is that meeting for example?

[mat.eo]

I’ve also noticed my assistants taking longer to respond.

OpenAI has been very silent about Assistants lately. Maybe an overhaul coming soon?

[niklasohrn139]

Hopefully, question is how soon

[excelformulabot.com]

this is unbearable

i’ve said it before - the assistant api is the red headed step child

no one at openai probably realizes

[PaulBellow]

I’ve passed it on to OpenAI. More soon, I hope.

[info1158]

Great to hear. Thanks! Hope we will get some answers soon then.

[aibnsamin]

Is it just down now? I can’t even get a response.

[renaudperretyvan]

Hi guys, I’m just sharing that I have the same issue.
I’m using GPT-4o Mini with function tools, and for the past three days, I’ve been experiencing huge latency—around 30 seconds more than before.

If anyone from OpenAI can take a look at our post to at least relate to the problem, I would appreciate it.

Thanks in advance.
A huge part of my business is based on this, so I hope for a quick answer!