I am PLUS user and this is getting painful, why are we paying for a service that is not working. It;s laggy, token responses are unfinished and short, keeps hanging, its so frustrating. In the mean time we are paying for a sub par service. I was willing to pay to use this but Im so close to canceling.
It definitely appears to work better once the USA has gone to bed. I noticed her quite slow this morning, but some queries this afternoon appear to be working quite well. Australian Eastern seaboard here.
Same for me. GPT-4 through API is absolutely useless – very slow and often “gateway timeout”. This is the case for more than one week, now. GPT-3.5-turbo and text-davinci-003 are working fine.
Same here. Attempted to minimize the max tokens but during the day in the US I always get a 30 second timeout.
Same here. And when it stops when sending a code is very frustrating. If you ask it to continue sending, the results are totally unreliable.
Hope they fix it soon…
Its still pretty slow, i hope they gonna fix this soon.
Same here. ChatCompletion requests that took 5-10 seconds ~2 weeks ago are now taking 30-40 seconds to complete for gpt-4 on a paid account. User experience went down the drain with such a delay.
Why is chat gpt 4 so bad at code now? It can’t remember what I prompted. 3.5 is better.
Has anyone had an update on GPT-4’s slow API response? Mine is getting worse and it’s a conundrum since 4 is soooo much better at what I am asking it to do.
I am using the API with a key I paid for
My response times are on the order of a minute for GPT-4. It depends on how much I give it to chew on.
Using it in completions mode/style (no context) I can get quicker responses for simple queries like: “write a limerick about a dog” take about thirty seconds
Refining it in a chat context quickly starts taking time and tokens
I am happy with it. Once I realised it was not timing out and learnt to wait.
11th of May and ChatGPT-4 has become almost unusable due to poor performance. I’m not sure I can justify $20 a month for a substandard service.
its really annoying. if I had been aware of this, i wouldn’t have upgraded. I upgraded due to the need for better performance but this is annoying!
Does anyone still experience errors with GPT4?
I just started paying for GPT 4 and it SUCKS!! Slow. Times out all the time. Doesn’t respond anything like as well or fast as the free version. Why I’m I paying for a beta version that is so full of bugs and barely works???
Here the same. Paying for it and its unusable.
Seconding all what has been said above. Actually, if there is not going to be a fix very soon for poor GPT4 generation speed I will be cancelling subscription. It’s borderline unusable for more demanding interactions that require lots of text to be generated, and GTP3.5 does it an order of magnitude faster. Is here someone from the core team to actually comment on this? Seems like paying members are talking to a wall here.
GPT-4 is an increased computation model by design. It was even slower at release and full intellectual power.
You can use your high-end consumer hardware to run an open-sourced language model at the size limit of what your computer can do - and get far lower token output per second even when it is the only thing running. Increase the cost per 2TB/640GB H100 inference server to around $250,000, and increase the cost of a rack instance of GPT4 to $1 mil that you get access to for a minute of supercomputer computation to answer about silly ASCII art.
Because it looks simple doesn’t mean it is simple.
Experiencing really slow output / response times on GPT-4. Has been the case on and off (mostly on) for the last 3 months. Has anyone been able to figure out if this is a browser level problem or if its OpenAI’s servers?
Still painfully slow on occasions for me. I changed tabs, carried on with some other work bits and then switched back to GPT after ~30s and it was still typing the response out, 1 character at a time.
Then 5 minutes later and it’s back to normal. Rinse and repeat.
It’s defintely them as per my previous comment.
It’s slow one minute, and then 5 minutes later it returns to normal, suggesting an overload on their end.