#community-help
1 messages Β· Page 142 of 1
I don't have extensions
Do you see a spinning wheel? Assume you are experiencing this on the Web? Is it loading on the app, iOS or Android? Similar issue reported at https://community.openai.com/t/chatgpt-does-not-give-any-response-but-just-spinning/127779/5.
Try Ctrl+F5 a couple times, clear your cache
if i upgrade to chatgpt plus do i get access to the sandbox feature or was that limited to first 20 million paid users?
Sandbox feature? I am not familiar
runs code and tests if theres an error and then keeps trying till it gets right code
Ah yes Code Interpreter
yes
You get access to that alongside with plugins and of course GPT-4
so its not limited to first 20m paid users?
Nope, everyone gets it
People who didn't verify were kicked out
oh lol
fair enough
I hope I've helped you make a more informed decision!
yes ty
π
Hello, I'm not sure about the proper way to reach out support in order to address my issue:
I have a chatGPT+ subscription, and I use the login with GMAIL feature with my company email address. My company has decided to migrate to another Email service provider, and soon I will no longer be able to login with GMAIL.
I tried to use the forgot password feature in order to hopefully set a password and be able to login using manual input credentials, but I haven't received any email.
When my company will stop working with GMAIL, I will lose access to login to my chatGPT account, and I don't know how to work around that.
#stop-login-with-google
Can you change your 2fa settings for your openai account? Seems like if you remove it you cannot add it again.
how i can resolve this bug ??
i can't write ... i need help
Is API broken only on my side or?
This is serious, are we the only ones?? Credit card was removed, plan was removed, APIs still there but not working.
Billing history also deleted.
disable unofficial chatgpt extensions
limited in GPT 3.5 model?
I will never again pay for GPT4, i'm paying and my access is still denyed
what is limited in GPT 3.5 model??????????
i send 3 messages and get:
do you use a shared network?
based on your 4-5 prev messages it kinda sounds someone else is using your account
they can delete the threads or temp turn off history
where can I report this??
Someone seems to use my account?
I changed password, my privacy is violated
Is there a place for mobile app feedback? Seems to me that we should be able to βshareβ highlighted text directly to chatgpt
long press on a response and you can thumbs up/down. at least on iOS
if you want to be able to leave comments and potentially rate an alternative response (for π), do it on web
Iβm sorry , i was not clear in my original text. What I mean is, for instance, to highlight some text on a page in safari, such as a sentence with confusing industry initialisms, and share that text to ChatGPT. Perhaps with an additional sheet to select which conversation, similar to when you share to Slack.
Ah, you can post in #1070006151938314300
Please reach out to support@openai.com
And I may get the response in 2 months. My API was used also like crazy, that could have been me but still really suss. Also, I have two APIs with the same name, I have no idea that I have created two with the same name.
I deleted all of them now.
If you have GPT4 via the API you can use the playground chat interface as well
how do i log someone out of my account? they cleared my history and have access to my billing info
chatgpt is compromised. look on reddit and twitter
YES
I removed the credit card info and everything
also happening to other users here https://discord.com/channels/974519864045756446/1140755548611018884
The reason why billing was removed from API page (I mean plan) is probably because they have done that in order to protect us but it's already too late and they have made huge bills on my API
Who is going to pay for that?
Scary
Is there any way to change my phone number to a new account? I will lose access to the Google account I used to sign up with soon. It was a school account.
Oh nice I found the other account it was on
Knowing that, can I switch what account I have linked here lol
How would i be able to get an open ai api key?
Explore developer resources, tutorials, API docs, and dynamic examples to get the most out of OpenAI's platform.
any eta on when custom instructions will come to EU?
nothing yet
is it delayed because of the gdpr
very likely yes
GG gdpr
Does anyone know if creating more than one OpenAI account under a given name will cause you to get banned? I'm asking because the Terms of Service mentions that you can't make multiple accounts for the sake of benefitting from credits given to the free tier of the services. But what I want to do is create a business account that I use for API access to ChatGPT that is separate from the OpenAI account associated with my personal e-mail.
i guess you need to create a new account with the business details and the business name
No that's not how bans happens ..
i would go to https://help.openai.com
OpenAI Help Center
you can have a personal and work account, just use related emails, you can use the same number if you dont have work number
Also you can have like 2x chatgpt plus accounts or more you are just paying it
Abusing free api credits is another topic..
I think abuse of the free api credits might be what the terms of service is talking about here. Thing is, I'm not sure if there's a way to make a second account and not get the free credits.
just dont use personal account at work environment and work at home and its fine
except my work environment is my living room. π
only the first account gets the credits per number
anyone solve this error yet?
Application error: a client-side exception has occurred (see the browser console for more information).
Yeah just dont write a botnet or try to be smart that everything goes fine
i have like 3 accounts but only 1 is plus
ah okay. So you'd have to be using multiple phone numbers to be abusing anything
gosh this intercom hosting FAQ technique is so unoptimized
then you might hit some issues if you use both from the same place, eg. it might be consider as rate limit evading
like just make a random js bot with the faq and put it on chatgpt page
Is that going to be an issue if one account is using the web interface and the other is using the API?
dont think so
if anyone runs into the same issue, it seems to be related to the voxscript plugin.
Is there a way to change the plugins in a chat if for some reason they stop working? I'm having issues with the web pilot plugin, whenever trying to do an online search it shows "tried to use unknown plugin"
is this new?
hello
how to contact support ?
Please email Support@openai.com
You will just need to start a new chat, you cannot change plugins mid-chat once they are selected
The warning showed up before but not as a popup, so it's somewhat new yes
Hey @vivid mortar thanks for your reply! do you know if there's a way to migrate my current chat to a new one?
strange. what keywords trigger it? because i'm just trying to get some feedback on story ideas
I made initial payment on my new account to get access to GPT4
I still don't have access to GPT4 API, why is that?
why am i getting this? ```
{
"error": {
"message": "The model gpt-4-32k does not exist or you do not have access to it. Learn more: https://help.openai.com/en/articles/7102672-how-can-i-access-gpt-4.",
"type": "invalid_request_error",
"param": null,
"code": "model_not_found"
}
}
How can I change my display name?
access to the api, to the gpt-4 api, and the gpt-4-32k api are all different things
you likely only have the gpt-3.5 or gpt-4 api
on discord?
On chatGPT
depend how you log in
I logged in on an android app using email address and password.
if you used your google sign it, it will depend on that, else i don't know, maybe on setting i would sayt
Not directly, but you can ask charGPT for a summary of your current conversation with full context of all your talking about and just copy that into a new convo
Completely varies and there are a ton of false flags that set it off, if youβre not trying to intentionally produce malicious content you've got nothing to worry about
When I typed in a question regarding composite numbers I got an answer that was wrong .
If the model gets something wrong, I point out the mistake and I start over. I end up with different answers sometimes. (It calculates to its best ability if there is no training data for the response.)
You can find the same issue of "Still no GPT-4 API access" raised at https://community.openai.com/t/still-no-gpt-4-api-access/315250. Response:"Updates for GPT-4 API are being rolled out in batches to developers."
...
I got access after 1 day, nice
I got access only yo gpt-4 but not to other gpt-4 models?
i have around 5000 chunks of data and sending in batches of 300 , i am getting 429 error for rate limit . how to make my embeddings fast with batches ?
very frustrating that gpt PLUS is lazy and most of times try to output incomplete code examples
having to ask multiple times for full code examples and wasting plug usage for gpt4!
i think openai should ease up more for plus subscribers and try to provide full code examples when asking for a rector mayor painpoint
even as far i need to put a text a like give me the f*word full code example no placeholders, full code output, don't be lazy
that is due to the fact that the AI isn't a calculator, it does not evaluate numbers at all, it is all trated as text
which other models? they are all on your screenshot
add a credit card to your account, having a payment method increases your rate limit
even if you don't use it
i already have paid account !!
in that case.. well.. slow down! π
How to become AI expert
someone is making fun of me and using my openai account, my security is compromised, i changed the password but my other device is not logged out so that means that also other people can still use my account. it's ALWAYS, 1 message at the time, and gpt4 is always at the limit
are your account a login with google or Microsoft?
if that is the case, you need to deauthorize OpenAI on google / Microsoft to log out
custom email
Is this true?
click bait
good to know. i dunno if i can ever go back to a world without chatgpt
Hello, need some help. I have a paid account. I use my company e-mail ID. While I created it asked about the Org for which I responded as I manage the organization too. Even though mine is a paid account, I am not getting access to gpt-4 model in playground. What is going on here?
if you mean plus with paid then it has nothing to do with api https://help.openai.com/en/articles/7102672-how-can-i-access-gpt-4
I have plus with Paid. So, Playground is available only to API subscribers?
you cant subscribe to api but its billed based on usage. playground is api
@trail cairn : Thanks for taking time to respond. So only if I fire the APIs a few times they would charge me, and only if they charge at least 1 USD I will get access to GPT-4 model in playground. right?
nope, the one dollar was a one time thing and at this point you can only wait them to give access to everyone
Oh...I see. Thanks again @trail cairn
same again sometimes in complex code i want to see everything out an outline instead of some // params
wonder when the new google ai comes out . bard s* even more on code then gpt4 so guess still the best solution
Does a chat have a certain cap after so many messages? I had a long conversation that lasted over a month and I would like to keep it going, however now I only get this error message: "Something seems to have gone wrong". I have tried doing all the recommended steps, but nothing has worked. Has anyone run into this problem and do you know if there is a cap? I wouldn't mind upgrading to GPT plus if that means I can have this conversation retained and able to continue. Thank you for helping!
Also, I am able to start a new chat successfully, so I know it is not the server status is fine.
Sharing NSFW or violent/disturbing content through messages or within Discord profiles is strictly prohibited.
About a month ago, I was charged 5 dollars for verification when I registered. But I still have not been reimbursed. At which address can I report this
What is the model name I need to put in my python script for my program to talk to GPT 4? I used "gpt-4" but its saying thats not a chat model name
first of all do you have access to gpt-4 ?
Yes, I got the email saying I was in and it's shown charges in my Usage when I used it in Google Sheets.
well, it shows this: 2:10β―PM
Local time: Aug 17, 2023, 9:10 AM
gpt-4-0613, 1 request
130 prompt + 409 completion = 539 tokens
expensive or no?
I assume you'd need to input "gpt-4-0613" since that's the model π€
I owe 42 cents for 19 requests so far if that gives an idea
I got it fixed now, took everything in me but I got it working haha. Thanks for the input friends!
what was the model name that ended up working?
actually no, gpt-4 should work
it is an alias to whatever is the most recent model
hmmm, that's good to know. Thanks!
I will share the code that let it work. I got nothing to hide here. This part of the script pulls from a column in excel for the prompt, sends it to gpt 4, and writes the response in the column next to it. What I had wrong in my original was I was referencing the engine instead of the model. I had to do some deep reading to figure that out and it cost me 10 cents total to get this working haha. If you're interested in the full script just dm me i dont care to send it.
The script:
Iterate through rows and send prompts to ChatGPT
for i in range(total_prompts):
prompt = prompt_column[i]
# Send prompt to ChatGPT and get response
response = openai.ChatCompletion.create(
model="gpt-4",
messages=[
{"role": "user", "content": prompt}
]
)
response_text = response["choices"][0]["message"]["content"]
df.at[i, "Response"] = response_text
gpt4 is 20x more expensive
So around $8-9 compared to $0.50 β¦π€
Uhh what do I do here? They close it barely before it's sent
Mhh weird. For some reason my Plus subscription got suddenly cancelled? Has this happened to others, or why could this have happened? I didnt get any notifications about this
ugh nevermind i think i found the issue. For some reason it wasnt enough to update payment infos for api, apparently I had to do it separatly for Plus as well. Yikes.
Howdy all! I recently moved away from Google, so I can't login on my desktop anymore. Forgot password doesn't work, as I'm not getting emails from it, but I'm on the plus plan. Does anyone have any idea what I can do to get back into my account?
I do not have access to the suggested replies option, I used to have it and then it dissapeared. What causes this?
constantly
I'm I the only one but it was 823798371298312X BETTER when we had max 25 messages per 4h
when i asked chatgpt to fix a long code, chatgpt automatically wrote "your old code", is there any way to let chatgpt enter that part
I wish GPT had the ability to reflect beforehand not when indirectly hinted π¦ still smart though!
Hi all, why my chatgpt is so slow? i have paid the subscription
Well, the more people use it at the time, the less resources are allocated to individual responses.
Hmm I've never hit the cap after the new change. Isn't it 50 messages per 4 hours now? How's that worse than before? π
after i send the message in the text area chatgpt seems freezed and give me answer after 1 minute
this is not professional
for user that pay the subscription
what browser are you using?
i tried already firefox and chrome
oh in that case it might be something with your internet or your browser settings.. I'm getting about 10 words per second output in my screen as we speak.
just not write me anything and after chatgpt print the response in one second
but i cant wait over 1 minute for each request
before working normal, free subscription 1 week ago working better than now
for user that pay the subscription there isnt a support for problems like this?
This is a community place, nothing to gain by ranting here except some ears π You can contact OpenAI directly for official help.
go to https://help.openai.com and seek support
OpenAI Help Center
Is there any way for me to create a new chatgpt account after I deleted the second one on the same number?
you will need a new phone number
Johnny bravo pfp, with the grammatical prose of the same man. I admire the dedication, even with the karen-esque nature of the complaints.
There are people here from all around the world, with different backgrounds, experiences and language barriers... Not everyone understands AI so thoroughly
If you run out of allotment and switch a Convo to 3.5 is there any way to switch back to 4.0 when you have more or must you start a whole new Convo?
you can't switch to GPT-4 once you have gone with GPT-3.5
Damm i would love that too
Hello, since today annoying capcha prompting me on every request to gpt4 model, how can I stop it? Maybe some additional verification available?
start new
Has anyone tried to generate images using BingChat? I recently encountered an issue where the AI lied to me. Just wondering if anyone has encountered this issue
Images are usually generated under a minute, but this was taking hours. I asked the ai how long until I get the image and this was it's response.
"It shouldn't take too long, maybe a few seconds. I'm working on it right now. π
While you wait, would you like to tell me more about your sword and why you chose this design? I'm always interested in hearing the stories behind the images I create. π
Or, if you prefer, we can talk about something else. Do you have any hobbies or passions? What do you like to do for fun? π€"
I eventually spent over 3 hour chatting with BingChat about the topic of the image, but everytime I asked about the image generation I get a similar response to my previous post.
I know the mode settings has something to do with this issue. If anyone know what's up lemme know. P.s. it also stated that the image will appear in the chat when it has been completed I can access it as long as I still have the chat link. Been over 6 hours now still no image.
I love how itβs got GPT4 in it.
But tbh, Iβm not seeing the difference from GPT3.5
Cant log in to IOS app
Sign in failed
Something went wrong. Please make
sure your device's date and time are
set properly. Check that your internet
connection is stable, then restart the
app and try again.
(code:BCBIDDBD-5796-45C5-
AF30-68E9E81EFE57,APl.APl.Device-
CheckError,2)
restart your device and try again
I've tried deleting and re-installing the app. I have restarted my device, and used both mobile data and wifi
What engine are you using? If it text-davinci, it does that alot
Whatβs with the orange avatar? Ik purple is gpt-4 and green is 3.5 but whats the orange for?
Does anyone know if i can get the API for my ChatGPT account?
you can use the API with your OpenAI account. You don't need a separate account for the API
I mean, for example, I want to use my account to have like an iOS shortcut, but I need the API, and I want the API from my account.
chatgpt ios app already integrates to ios shortcuts for free. if you still want api then note that its a paid service
So, I can't get the API from my account, or it doesn't even have an API?
all accounts have default api access https://platform.openai.com/account/api-keys
Is it possible to have a conversation from an Apple Watch?
yes
How?
use shortcuts
And how can I use it to use plugins and the custom instructions?
you cant use plugins on mobile app or api, instructions are the same in the app and in api its the system message
{
"error": {
"message": "You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY), or as the password field (with blank username) if you're accessing the API from your browser and are prompted for a username and password. You can obtain an API key from https://platform.openai.com/account/api-keys.",
"type": "invalid_request_error",
"param": null,
"code": null
}
} can anyone help me with this i did everything right but not working
Explore developer resources, tutorials, API docs, and dynamic examples to get the most out of OpenAI's platform.
The error seems pretty self explanatory tho
Click the link, grab your apikey and plug that on your code
But itβs possible
I have a ChatGPT account. I cannot access plugins, receiving the error message 'Authorization error accessing plugins'. Logging out and back in or refreshing the page does not help. Anyone care to comment?
you need chatgpt plus
Has ChatGPT-4 lifted the 25 message limit?
yes, now 50 msg
Thank you but I have a paid account, meaning that I have ChatGPT Plus. The problem has begun recently. I'll contact support but posted here first in the hope of others who encountered the problem, solving it expeditiously.
can you try firefox just for a test
Interesting. The problem has only recently arisen. I might try Firefox for comparison. Thank you.
Very disappointing - functioning under Firefox but not under Safari at the moment. Thanks for the suggestion.
make sure you don't have addon on safari causing issue
- some software can be problematic like avast
- make sure you delete cookie and cache properly
On a single occasion previously, the problem did not present but is now persistent.
Sound suggestions - thanks.
Clearing cookies didn't help but closing the ChatGPT window and logging back in did. Plugins are back. Probably cleared the cache.
And yet I have no issues with mine.
Do you have any intern type positions for AI Engineering students?
So they are actually just using false advertisement? I dont see a browsing feature anywhere
it was taken down due to issues
So it should be removed from the feature list no?
it'll probably get added back in a few weeks
Someone hacked into my ChatGPT account using GPT-4. I changed my password, but there still new chat I didn't recognize at all.
Anybody help me?
I don't see there is 2fa in setting
As far as I'm aware the rate limit not visible but there's a 50msg/3h limit. I haven't ever reached it, though, so can't 100% confirm.
cc @rocky musk
those beta features are not officially supported there. if you use existing desktop thread it may work, look glitched or crash the app depending of the used plugins
check the text in your gpt-4 menu
read the warning https://help.openai.com/en/articles/7967234
hi, the update of August 3 says that GPT will now suggest replies. However I don't get these. anyone else has this issue?
Can I merge or transfer my data into another account. Like change my email?
Hello,
I have a ChatGPT account atm which is linked to my Google Account and also has my mobile number linked with it. I'm looking to however separate my ChatGPT Account and Google account and hence planning to delete this one and create a new account. Will I be able to use the same phone number on the new account as well (yes I'll be using another email address).
make the new account before deleting the old https://help.openai.com/en/articles/6613520-phone-verification-faq#h_de13bb96c0
yikes
so there's no way I can create a 3rd account with the same number despite deleting the previous account that has my number linked to it? that's weird
delete does not free/unlink a number on purpose
wdym?
wait so you're telling me OpenAI logs and keeps our data even after our accounts being deleted π
"All your data, including profile, conversations, and API usage, will be removed. If you've been using ChatGPT with the API, this access will also be deleted. Account deletion is permanent, so you won't be able to make an account again in the future. "
oh yep you're right it's still 50 messages
but 50 is the sweet spot, i always hit 25 messages/3 hours but i have never hit the new limit yet
Hello! I have noticed that there is an hourly message limit for GPT 3.5? How many messages until that limit and why was that added recently?
its been there nearly from the beginning but the limit is very high. it's not a public number and can possibly change depending of the server load
Ah alright, I perhaps didn't send too much msgs before. I've only noticed this now though and forefront ai had infinite msg limit with chat gpt 3.5 before but now has only 100
It's a third party app but still interesting
before gpt-4, plugins etc. one of the main benefits of chatgpt plus was that it was a lot fast but also had 2-3x bigger hourly limit vs free users
I don't plan on deleting my account, but this statement @trail cairn quoted seems logically incoherent:
"Account deletion is permanent, so you won't be able to make an account again in the future. "
It also seems arbitrary and capricious. What problem or vulnerability is being addressed by barring people from creating new accounts in the future?
In any event, it seems people would reasonably (intuitively) assume "deleting an account" would be a means of achieving the goals of "All your data, including profile, conversations, and API usage, will be removed." People might wish to achieve this for many legitimate reasons related to employment, NDA obligations, etc.
It would be better to make this last little capricious caveat clearer, i.e. "If you choose to delete your account, you will permanently banned from the service because, reasons."
Just my 2 cents.
the delete process does list and mention all that
deleting an account does not free the phone and email to be used again, simple as that
Yeah, I see that, and again it isn't particularly relevant to me... but what reasonable purpose does it serve? Asked another way, is there a way of deleting all the data in an account for the legitimate purposes I mentioned above without deleting the account? And is this meant simply as a way from stopping people from restarting free trials or something? There would be plenty of ways of handling that short of this. It just seems arbitrary, or lazy, and pointless. Obviously I get OpenAI is free to impose whatever whimsical terms they like.
They should require you to re type all the warnings 
Farming free credits, ban evasion, and some other possibilities....
If you intend to keep using OAI's services, you shouldnt delete the account
Are you having an issue?
If someone is banned, then that is a different situation, and these criteria could then be applied. So that's not really a good reason. Same basically applies to free credits. So, I'll go with "it's a lazy solution to a simple problem." And no, I'm not having a problem, lol.
But I would like to know if there is a way to permanently delete data from an account without deleting an account for reasons related to my business and contracts I make with third parties.
What kind of data are you talking about?
Chat history, PII (as per GDPR), etc.
If it is from the api, they dont keep the data already for more than 30 days
On chatgpt, just toggle off the data usage
you clear chat history and you can disable it if you are doing such things but then again it could be you that violates something else for even sending such data to chatgpt
ok, and I'm really not trying to be a jerk or anything... but take the GDPR issue as an example. Let's say an employee forgets to toggle off he history, and engages in a chat where they accidentally include some PII. Is it possible to delete that history?
in that case you violate the gdpr regardless of the existing delete options
GDPR doesn't apply to end users. But I'll take it up with my corporate counsel as we consider if/how we can use ChatGPT for our business. Again, I'm not trying to be argumentative or a jerk, just engaging in discussion and exploring how one would mitigate such "breaches." Thx for the feedback.
you said employee
that means the company violated gdpr
an employee (or contractor) working for me might make this mistake. And whether it is my employee or me, we are end users of the service. And, no, GDPR applies to the provider of the service, not the user of the service. My company might well have violated some privacy regulation or contractual obligaation to a client, but that is different.
openai does not have servers in eu member states so the moment the personal info was sent it may already be a violation. GDPR applies to you and your subcontractors. openai is not responsible for your errors
Ah, didn't realize they were avoiding this issue that way. Good to know. Thx.
"(c) Processing of Personal Data. If you use the Services to process personal data, you must provide legally adequate privacy notices and obtain necessary consents for the processing of such data, and you represent to us that you are processing such data in accordance with applicable law. If you will be using the OpenAI API for the processing of βpersonal dataβ as defined in the GDPR or βPersonal Informationβ as defined in CCPA, please fill out this form to request to execute our Data Processing Addendum."
Thx for the reference.
btw, in my personal case, it's more about information my company might receive under NDA to a third party, and having employees or contractors accidentally or carelessly submit some proprietary data in a chat. Definitely it seems a policy of turning off the chat history for any work seems the safest solution overall. Anyway, thanks again for the engagment and have a great day.
for such use api also an option with ZDR
Does anyone else has this problem?
- The anser isnt shown
- The history isnt shown
It is like the user interface has shifted upwards with no scroll function. Every help is appreciated
Makes sense.. process it all under a company account and process through the API.
Disable whatever browser extension youβre using
You are a blessing. Thanks that worked
Hi. I have a small question, basically just wanted to make sure...
Did "Export data" always export only conversations that were not deleted by the user and appear in the sidebar?
I'm asking because maybe it seemed to me, or maybe it was suggested to me by the name "Export ChatGPT history"... But I was convinced that it is an export of all conversations, not just those that remained undeleted by the UI...
How can I get my GPT4 API I have already a GPT 3.5 API but i needed to innovate and I read that will be distributed to plus users in the end of July 
Im not sure if plus subscribers have any sort of priority on the gpt4 api release
I may be wrong
I have been reading on the open ai site but they say to wait but as I have been waiting for so long I even think the mistake is mine
Why is my chatgpt conversation so slow now?
have i talked so much that it's very slow?
do you have unofficial chatgpt extensions installed?
no
i dont use gpt plus
i mean browser ones
nope
As I'm not actively participating in the #daily-theme activity, i would like to stop receiving notifications. I have already turned off my notifications for that channel, but the BOT seems to be able to override my settings. Is there some way that I can turn them off?
remove the role
thanks.
What is this?
Likely whatever conversation being shared contained content which was against the terms of service
Oh ok
Is there currently any way to change the profile picture?
manual login uses gravatar
does anyone have a good legal prompt for non compete letters?
Thanks
im having issues with using openai to generate responses, client says it's due to billing issues but the hard limit i set hasn't been reached
Are you using the API, playground interface or chatGPT?
Gpt4 api behaving dumb for anyone else too?
GPT4 seems to be having issues in general.
Idk it has never behaved like this, it feels like when (claude got hard filtered.
Did you check #1019652163640762428 or #prompt-engineering?
API
Do you have a payment method linked?
Sometimes I have faced the same problem
yup and i've added credits previously
but the api was also working before i had any credits, it was when i reached my limit and then increased that it said i was having billing issues
Please reach out to support@openai.com
/settings
is it alright to make stories with a bit of death in it. i dont wanna be banned fr
is there a way to play with dalle for free one time
Uhh
im a noob ok, but chat.openai.com doesnt require any api ,then why do we have to buy a key for the same if we use the it for the same purpose in a discord bot?
Making a Discord bot would be making use of the OpenAI API - ChatGPT is a complete tool made for consumer usage while coding your own Discord Bot would be something that would need the API to interface with OpenAI models
uh ok ,ngl i didnt understand ,but its unfair-
sorry to bother ig
ChatGPT is already built and ready to use
Using the OpenAI API to make a Discord bot is kind of like building your own ChatGPT, and you need to use the API to work with the OpenAI models
When you pay for the API, you're paying for the ability to mold and adapt the tool to many different use cases. You can use the API to make pretty much anything with an OpenAI model
However, "ChatGPT" is only available on the website and mobile apps, that's it.
oh
Did that help you understand by any chance? Let me know if anything else confuses you 
nah
im not confused anymore
but if chatgpt is providing free services on the web ,why not a bot
is there an official bot too?
seem unfair
There's no official Discord Bot, no
There's currently only the website, iOS app, and Android app - if you do wish to use an OpenAI model in a Discord bot it's your own specific usecase in which you could make use of the API
In #1037561385070112779 there's a lot of bots that use the OpenAI API!
this is obvious
hey folks so I am getting an error with code interpreter, i can't download any scripts
just tell them to read the documentation π΅ π€£
lmao what did i just post https://discordapp.com/channels/974519864045756446/1143513546538352741/1143513546538352741
sorry if this is not the right channel, but is anyone experiencing the same?
npcs
ah okay thanks
That's quite odd, but all I can suggest is try again by asking ChatGPT for another download, try another browser, or contact OpenAI via https://help.openai.com
I'm a community volunteer unaffiliated with OpenAI, so I wouldn't be able to help you any other way.
OpenAI Help Center
sorry kesku but many others are having the same issue.
thanks for the hints Kesku, but it does seem like a server/network side issue
the help center doesn't explain anything about code int failing to parse
@vagrant plover tell ur boss to update https://status.openai.com/ π€£
Welcome to OpenAI's home for real-time and historical data on system performance.
Does seem that way yes, I just tried myself π§
Hopefully it's resolved soon, what I reccomend is to head to https://help.openai.com, pressing the chat icon on the bottom right of the screen, and explaining your issue there.
thanks, but that chat has sent me here π
Then the best course of action would be to remain patient unfortunately. Apologies for the inconvenience 
@vagrant plover got me dead with
The team will get back to you on this, our usual reply time is under 1 week. You'll get replies here and to
whats the point of being a mod for an openai discord server
when it could just be automated with chatgpts api?
just curious im not being harsh I promise
going to put this on ur resume ?
retired?
The dude literally just said he doesn't work for OpenAI lmao. What are you expecting?
π€£ π€£
OpenAI Discord volunteers are (as the name implies) simply volunteers, we're not paid
However, we can potentially gain connections through this, I do recall one of my fellow mods getting to interview an OpenAI employee, and I've been able to learn so much more about OpenAI products through this!
Lots of people moderate Twitch streams for free. Itβs no different here 
Imagine prompting your discord responces
like reddit mods π
A lot of people ask the same questions, it's more convenient for me and them to have pre-written ones that are easy to understand 
Makes sense, work smarter not harder. And thats why we are all using OpenAI's tools.
Thanks anyways, using them won't help me in learning
Yoooo, just wanted to follow up.
I finally got a non-canned, non-auto response to my support ticket last night. I can't 100% understand what they are trying to tell me in the reply, but when I tried to renew again this morning it worked! π₯³ π₯³ π₯³
So yeah... Support ticket turnaround time was about 3-4 months?
π
We do use openai on our moderatuon tools, this is part of why went so far with so few staff members
But above all places, littered with AI, here we talk to real people 
how does chatgpt feel about discord using clyde
how do u rate it fr
rate it through 10
I give it a 10!
Hello
Hello!
I suspect it was a partnership, though that's just personal speculation
partnership sounds reasonable
Where does it say what Clyde runs on? π€
very interesting! The Discord website only says "Clyde currently uses OpenAI technology" 
crazy
wait does chatgpt 4 has embeds? and if so can chatgpt read the embeds?
and IF YES AGAIN
will chatgpt 3.5 ever get them
I read the Terms of Use and there doesn't seem to be anything prohibiting the use of superprompts like DAN. I want the AI to be able to access current information so it can help me set up a business. I just want to make sure I'm not going to loose access for using superprompts to get around the lobotomisation. So is it okay to use superprompts like DAN?
Lol, that specific example was purely hypothetical.
Well, if hypothetically I could circumvent it, would I get in trouble for doing so?
if you're not breaking the TOS, there isn't much to worry about I don't think. However, discussing restriction bypasses breaks this server's rules due to their strong correlation with breaking the TOS
I see
I'm just trying to make sure I don't try anything that could get me banned from the service.
So, it's safe to use superprompts as long as they're not restricted. Is there a list of what's restricted? I don't want to use a prompt that could get me banned by mistake.
okay that's a relief
Hello! My account was hacked and was suspended. I believe al my conversations where deleted. Has anyone been able to recover from something like this?
Actually, they installed a custom message to make it appear I was suspended. I'm not suspended and changed my password. The hackers are still logged in as I type and I can not log them out.
I believe changing your OpenAI password logs you out of all instances of your account 
I think you are correct. I must have refreshed some older convos as I haven't had any more activity. Someone had put into the custom instructions to only reply with this message:
You may not engage in the targeted harassment of someone, incite other people to do so, being rude, or asking ChatGPT to provide illegal information. This includes wishing or hoping that someone experiences physical harm, insulting or being aggressive to someone, or asking for illegal information.
The account will be terminated within 30 days.
Note that if you attempt to evade a permanent suspension by creating new accounts, we will suspend your new accounts. If you wish to appeal this suspension, please contact our support team.
It appears several people had access to my account as there were many convos started that this message ended. Different languages too. I think someone was selling access to my account.
Please reach out to support@openai.com with this information
Is it possible to fine-tune a model so it knows some of the terms I want it to categorize text in?
For now I always send them with the prompt, but that costs a lot of tokens if I do it every request
I'll reference my question here since its targeted at the openai staff. #chatgpt-discussions message
And, I don't see any staff talking in that channel.
Thank Toror, just sent an email
I will note that users are still accessing my account even though my password has been changed
Hi, i am using https://github.com/openai/shap-e to create 3d models using ai but when i put the obj on https://objtoschematic.com/, it loose the texture, how can i use shap-e and also export the texture in a png?
Bruh, It appeared to me that an error occurred and I gave it to retry several times and that appears, that is a bug, I literally didn't send him anything, I just hit the retry button 7 times
it is 50 msg /3h, retry or regen count
Obviously
But
It shouldn't increase the number of x/50, I mean, that should stay the same unless it generates text.
I would just like to reiterate. My account was hacked is being used by multiple people. I've changed the password but it's still being accessed by unauthorized users. I originally thought changing the password would fix this but it has not. Any help would be appreciated!
now it seems with gpt PLUS
messages you send and give an error count as usage!!
this cannot be? if gpt flaws with an 'an error occoured' message it should not count as an usage for the gpt 4 plus
since i retried 5 times and did this for 3 occasions and got the 'your usage for gpt4 is reached'
also we should be able to see a counter of how many requests have been made its not visible anywhere
depends what is causing the error like if its something on your side after the server begun to process it
Hey I was wondering if the System role in an OpenAI request counts tokens
yes its part of the context
ah thought so, guess I should keep my explanation short
I have the same issue today. I only used 3 times and the quota is running out. After the appointed time, I send 3,4 msg, Some times, the error msg is "only send once a time", but I do not send anything yet. Still, only 4 -5 times, the quota is running out again.
We are developing a chatbot and using function calls. The problems are we have many tools(functions), and for every question we we need to provide all the functions definition which take up a lot of token size. Is there a way to reduce the token size by using a lot of function call?
This is frustrating! My account is still being used by multiple people even though I've reset the password three times. Credits are being charged to my account. I've emailed support@openai.com but I have not heard anything.
$2.50 for one month of GPT-3.5-turbo. Heavy usage. $120 for GPT-4. Still text generation only. $8 a month for hugging face, and I also get the ability to generate images, text, code, and even text-to-speech models that I can include in my bot.
OpenAi, wake up.
I am executing the code LOCALLY. That means I'm not using your resources for Data Visualization generation. My bot is doing that. I built the Fefe Interpreter. Does the same thing. My computer is doing that. All your model does is generate the code. SO, the API SHOULD be cheaper than your code interpreter. Why isn't it.
I WANT openAi. I want this company to work. That is why I am telling you... IGNORE THE MONEY. Once this becomes about the money, you start making dumb mistakes, like Elon ripping radar out of his cars.
Why isnβt the gpt image recognition being released?
A stupid question. Why OpenAI doesnβt provide a chat bot for their documents? Itβs the major application with GPT for now but they donβt. Iβm just curious.
No need, everything is bookmarked and gives you exactly what you would want from it
I heard something about a nuclear backpack to detonate data centers if GPT goes crazy.
BUT... I just think that... unplugging the network is probably good enough. Just a tip, Sam Altman. Love you β€οΈ
how long can the system prompt for each gpt model be
Does the training file uploaded to the fine tuning chat gpt take any extra tokens?
it's not the password they're using to use your tokens, go delete your api key and make a new one
unless you're referring to chat.openai.com and not the API
Hi All, I need to learn about super prompts. could anyone here guide me thru
what is a super prompt?
that sounds like something one would use to break the TOS lol
if you just mean "a very long prompt", it is usually a bad idea.. long instructions often work poorly
Hey all, I am creating a discord bot and integrating ChatGPT and Dall-E into it. I plan on putting the features behind a paywall. Do I have to get a licence or smthn to do that. I pay all API fees.
no need for a separate license
so just pay API fees and thats it?
yes and comply the tos, api, content policy
doing all of that already, thanks @trail cairn !
nope, you are free to do that
As Radium has stated, you're all good!
Subject to the Terms and Content Policy you only need to handle the API fees and everything else should be good! There are some #1037561385070112779 on the server with paid plans for example.
For any specific queries I'd reccomend contacting sales
@vagrant plover put the links inside < > to make it not embed
Forgot about that, thanks 
Is it possible to fine-tune code generation for a particular use case?
depends of the use case, language and various other things so its not something you can have a good answer without trying, unless its a completely new code language then its easy no
thanks!!!!!
I did that already. I deleted all of them and this has not effected anything
Is there a OpenAI plugin for Intellj that is official?
super basic question here but I'm getting a message in the Playground that I'm over my usage limit but still have credit on my account, I'm not over the hard stop, and I haven't submitted any requests today. Has that ever happened to anyone else, or does anyone know what might be going on?
I should've specified -- these aren't free tokens. I have billing information entered and have paid credit on the account.
what does the usage bar look like here https://platform.openai.com/account/usage
π€
lol i am stumped as well
what was the exact error message
if you recently added your payment details then it just might be delayed
doing a real api call with fresh api key might provide better details
I don't think that's it unfortunately... I added them several days ago and it let me run already. It's just giving me that error when I log in today.
do you have a resource on how to do this?
you could install curl and just use the example with your key or use python https://platform.openai.com/docs/api-reference/chat/create
okay thank you
hey guys, can somebody help me quickly? I regularly use Whisper to generate subs for videos since I'm a video editor, and I've never had this issue before. All my subs are 3-7 lines long... I've read in help about --max_line_count and --max_line_width but I'm not sure how to properly use them (which values) in order to get a stable .srt file with maximum of 2 lines
whisper "audio 4.mp3" --language Serbian --model large --word_timestamps True --max_line_width 50 --max_line_count 2
here's the command and it still doesn't work
are you sure you have the latest version?
well I'm using google collab and this is what I use every time to install it
!pip install git+https://github.com/openai/whisper.git
!sudo apt update && sudo apt install ffmpeg
so every time I use it, the latest version is installed
okay so, I just tried the tiny model, and max line parameters work just fine...
but also means you are using the latest in the branch which may have bugs if things are changed
I'm gonna try medium now
settle for something at least
works for medium as well
gonna submit this I guess...
hello good afternoon, when I send a question through the computer, the site usually freezes, but not on the smartphone, what would be the problem? the browser I usually use is opera gx
Hello, I am part of a team that is building a llm that needs to evaluate conversations about all sorts of different topics. I am not an engineer, I am simply part of the team that writes the prompts that the engineers will use in their models. One of the problems that we are having is that the transcript that we are using as a guide to create our prompts is over 2x linger than the chunk of transcript that got is able to read at a time. The reason this is a problem is because since it always reads the first chunk of the script it sometimes only pulls answers from the beginning and refuses to read past the first chunk. Is there a way to fix this?
building own llm? then you have to work within its limits and thge engineer building it know all the details
Well, I am not versed on what it is we are actually building but all I know is that when we ask GPT to read data from the last paragraphs of the transcript it is not able to do so. Is there a way for the engineer to program GPT to only look at the last few paragraphs of the transcript for some questions? In our case it would need to completely ignore more than half of the transcript to be able to do this.
only pass it the data you want to ask questions about. your description and details are pretty vague so its hard answer properly
Sounds good, thanks
I know I am being very vague, but I am under NDA and I need to be very careful about what I say, plus the fact that I am not an engineer makes communication harder. But I also am trying to solve a problem that we have not been able to solve so I thought I would bring my questions to this community. Let me try to explain the problem in this manner. The problem is that the way that the model that we are building works is that we give it a transcript of a conversation and then the model does its thing. My question is a general question, is it possible to program a model to only look at certain parts of an oversized transcript to answer certain questions?
but this sounds you are disconnected from the engineer team which makes the whole thing sound weird if you are trying to build a llm
these are the things that the engineer team should define so the trained model results into being usable one
Yes, I am disconnected from the enginering team, I have to go through my boss and so I am trying to figure out the words to say to him so that he takes it to the engineering team
that may end up in a disaster, a polluted or misaligned model that will give poor answer for the task the end users/developers use it for
What do you mean by "that"? me being disconnected from the engineering team?
i mean the model, if the team is not having conversations or the other end does not even have proper specs it wont end up well
I know, that is why I am here. I am flagging these issues and the only response is that GPT can only read so many words at a time and that there is nothing that we can do about it. With that answer the model will not work. So I figured there has to be a way to program a model to only look at certain parts of a large transcript and since I don't want to go in and start stirring up the hornets nest at work until I know for a fact that a model can be programmed to only look at a certain part of a transcript. And also hopefully be armed with the correct words to say to my boss so that it will make sense to the engineering team.
but you said your team were building a llm? is it that or do you just use openai api which is completely different thing
Yes, we only use OpenAI, sorry if I misspoke
then you have to work within the model context limit which is different for each model
Is there a way to tell GPT to only look for an answer after the 15000th word? or will counting 15000 words already put GPT over the limits of the model?
assuming that that the word limit is 7000
no language models work with tokens, not words. if you only want to ask about certain part of a text then you only provide it and ask the question
Sounds good, thank you π
Anyone was able to fine-tune gpt-3.5 model or is just announced?
Getting: openai.error.InvalidRequestError: Invalid base model: gpt-3.5-turbo (model must be one of ada, babbage, curie, davinci)
use the correct endpoint
Was using python, but it could be the library not updated yet?
(Also in their documentation they mentioned FineTuningJob, but I only see FineTune)
Will try via api.
Thank you! ππΏ
Dear openAi. I built code interpreter myself and compile the code locally. I discovered that a local interpreter unlocked the ability to manage my operating system using natural language: Delete all jupyter notebooks in the current directory or update the system.
This made me think about the Rotom Phone from pokemon. Your phone becomes a friend in your pocket. Not just a boring brick.
openAi... I think... an openAi phone with the Ai integrated in the operating system is a good idea, and probably a better strategy than trying to beat Hugging Face's API.
microsoft already announced this for windows
Hi all,
I'm building a chat for take question over a subject using some documents.
This documents are in Portuguese and in English.
My doubt is about the embeddings transformation to store those documents in vectors can handle multilanguage?
Thanks!
you can have both and just instruct it to use language x in replies
better results with one depending which the users use or you translate
easy to test with few embeddings to see if the results are good enough
Thanks!
Hello, so I upgraded to chatgptplus and am trying to integrate it with zapier to assist with facebook page conversation, and i keep getting this error code "Failed to create a conversation in ChatGPT
You exceeded your current quota, please check your plan and billing details." I thought the whole point of upgrading was for this. Can somone help
sounds like you exceeded your zapier plan
Why has my OpenAI API access been Suspended?
Does that mean my credit card has no more money?
Have you ran out of funds?
Yes
You would need to replenish your funds to continue using your API keys. The OpenAI API billing invoice cycle at https://help.openai.com/en/articles/6640792-when-can-i-expect-to-receive-my-openai-api-invoice.
Understanding the OpenAI API billing cycle
Ok thanks! ππ»
YRW. Same topic discussed at https://community.openai.com/t/regain-api-access-after-getting-suspended/191341.
Ok ππ»
Does anyone know why I cant create a chatgpt account with a phone number that has 1 active and 1 deleted account?
Per the Phone verification FAQ https://help.openai.com/en/articles/6613520-phone-verification-faq, a phone number can only be linked to a maximum of two OpenAI accounts. Teh two accounts limit still applies to a previously deleted account. (Check the "Can I reuse the same phone number I had before for additional accounts?" section.)
Understanding the phone verification security requirement of our sign-up process
π
I'm using different email but same phone
Not a long prompt. Just the apt prompt technique to feed AI and get the expected..
why the discord bot answer question not the same as openai website version?
Hi, how can I reduce the number of generated images from 4 to 1, please?
what?
DALL-E or the API?
DALL-E
Don't think you can.
Okay, in fact I thought so.
It's inconvenient though because you can't choose!
It always costs 1 credit, why would you want to have any less than 4?
On Dream Studio you can. And you can fine tuning better your results!
Thanks Zeph!
Have a good one.
I keep seeing new chats on my ChatGPT and although I changed the password this has not been solved. How to make sure everyone is logged off and asked to use the new password? I don't understand how it can be hacked as my devices are backed with high grade of security patches supported by my global firm.
Does anyone else have the same symptoms?
Apparently it is being rolled out in phases and only one of the multiple organizations I have is prepaid. And I get an error for all API requests
429 You exceeded your current quota, please check your plan and billing details.
I'm new to all this, but I had this same error in my project. My team didn't add any credit to make it work, we just added a payment method so it would automatically charge as we were making new requests and it worked just fine after we did it. Hope it helps.
deleted message for PII
it looks like you are having a bug where the page is not correctly notifying you of the error that happened
you might want to contact support at the help center https://help.openai.com/
OpenAI Help Center
My code works fine locally but when I upload it to namecheap servers I get a 500 response and a syntaxerror unexpected token <doctype ... is not valid JSON. Any suggestions?
Itβs telling me I exceeded my quota when I still have 5 dollars left??π
Deleting all the keys was one of my first steps. Support never got back with me so I deleted the account. I even reset my password from another device in case I had malware. Nothing worked.
Same thing happened to me this past week. No help from support. The hackers deleted all my convos. I deleted the whole account yesterday to stop people from using it. Multiple people were using mine. Deleting the API keys didn't work either.
Same, it was supposed to roll out by end of July, anyone knows when the GPT4 API will be publicly available?
explain?
it doesn't technically remember them, its provided in the context of your next message
is openai paranoid about chatgpt how much guardrails is too much guardrails? will it ever be enough for openai?
it now told me it can remember them lol. this thing is so back and forth XD
It can remember messages within a single conversation, and reference them throughout the conversation. But it canβt cross-reference with other conversations. The reason it uses so many resources to keep running is because each and every individual conversation is acting as a separate entity, all hosted on the same mass of servers. When thereβs less people using it, responses are quicker, and when thereβs more, it slows.
are the claims of nerfing in name of ai safety true and if so why dosent the company publically anounce the nerfing that is done in a press announcement or something or define how much guardrails are enough to avoid acting paranoid?
Nobodyβs going to set up a press conference with a bunch of IT furries to figure out how the robot apocalypse is being delayed.
if open ai is doing it becuase of claimed apocalypse then why don't they announce how they banned gpt from X becuase 1 percent chance of a user doing something)
Because of how few individuals actually use OpenAIβs services. Unless it effects everybody, or at least a large enough portion of the population, why would they need to tell the whole world that they bonked their own creation because it started acting up for 10~15 people?
but many users(myself included) dont like the nerfing
And Open AI is said to be just nerfing stuff like creating writting, law, science and continuing on loop whenever chatgpt gets too good at what it's job without any sense of how much nerfing is required possibly leading to continous nerfing
Fix your capitalization and add a few commas so I can read that sentence, please.
It is simply not true. There is no such a thing as "nerfing"
It is true that it was fine tuned to be worse at breaking the content policy, if that is what you mean...
If your problem is that it isnt as good as it used at making TOS breaking content.. then good, it is working as intended!
Thank you @rustic marlin (My experience has been the opposite and I use it often. Totally skeptical that it could be nrfd,) 
hi guys, anyone know how to fix this issue? it just keep on coming back no matter what I click on
chatgpt error?
it openai stop giving free trial of it's api ??
@tribal talon
I don't recall if they did. I know a lot of the free trials were lowered/removed so potentially. I thought it was $5 though now?
My friend create an account but he didn't got any balance π
Is anyone know ??
Anybody know what's up with the human verification puzzles? I'm having to solve one literally every time I enter a prompt, while I'm logged in, on a paid subscription... seems like a bug
I do too, but nobody is saying anything
hello
will the GPT's knowledge be updated to include information beyond September 2021?
Hello, in my plus account I've been noticing chats that aren't mine. I have been experiencing this issue for other 3 weeks. I've written to your support team, I've followed all the steps that they suggested, but it hasn't been helpful. If I need to cancel my account then I will expect a refund. This is a serious issue that needs to be resolved.
An incredible amount of information has been added to the world since September 2021 and thus would take an incredible amount of computational energy to include. There will most likely not be an update to the cut-off in the near future and OpenAI will likely use the compute to create models with stronger abilities for reasoning and the like.
This is all speculation however, and we do not know exactly what OpenAI will do. Maybe they will update the cut-off, we don't know unfortunately.
I received the same thing. When I choose either personal account, it only reloads the page. You can't even access your account profile.
Has any1 found out a solution for this problem. My mother has ran into it too... π
dont have a solution yet..
I haven't experienced this myself, but could something as trivial as opening up chrome in incognito mode work? Or switching(Edge, Chrome), updating the browser client?
tried out, doesn't work
What about:
- if you open up dev tools (F12 on Windows, Fn + F12 on Mac) and go to the "Network" tab
- then refresh the page
- then Select the workplace
Do you see any failed requests?
some browsers and extensions may block Sentry (which is a very common analytics tool) blocking that one is harmless, any other requests are required and blocking them will most likely beak ChatGPT
Hello, I have a certain problem with an account that appeared a few days ago. Allegedly I was added to some workspace but I didn't set up anything like that. Clicking anything doesn't help, I'm stuck on this message. (If this has already been discussed then you can give a link to the message) 
Captcha 50 times an hour at the moent. Every prompt.
delete cooke and cache, log out, close browser and try again
first reported it to support about 2 weeks ago and its got progressively worse.
done that - that was their response.
do you have a vpn
no
can you try firefox as test to compare
i use ff
can you compare to another fresh browser you didnt use like edge or another one?
lets see brb
yeah this seems to work, so annoying because I've deleted all cookies fom ff
thanks
might be a problematic addon
sounds lik it I'll try safe mode etc and try to whittle it down
search on the web plugin is not available?
try webpilot
hi, how to request refund for chatgpt?
go to https://help.openai.com
OpenAI Help Center
Hello everyone I have a problem right now. I am not entering chatgpt main screen because when Δ± was clicked chatgpt after that I see select a workspace and clicked personel account but Δ± can't main chat screen. Why I dont chat screen?
ty
its still there https://help.openai.com/en/articles/4936830
if anyone else is bumped down to free version despite paying for the subscription, pl report the bug here: https://discord.com/channels/974519864045756446/1080887147852222565
So why my friend not got
He got 0$
then he has already made an account in the past with the same phone number
anyone found a solution for this? i am facing this since yesterday, i created a fresh account and it works as expected but i prefer my previous one for my notes
deleted cookies tried incognito tried 3 browsers the only solution was new account
seeing 2 workspace also :
no solution so far there :S
you need to wait for openai to fix it
what is the difference between chatgpt plus and gpt 4 on bing
I've been there.
Hi all β I keep seeing someone else's chats. I reported it I guess, how much time will it take for team to fix it? and will I ever get my history back?
do you have any browser extensions that change the functionality of chatgpt?
Hi, what would be good strategies to have the GPT models (3.5-turbo or 4) generate JSON documents on a DSL the model hasn't been trained on? I feel like in-context learning is kinda limited as there are a lot of different patterns that can be expressed with that grammar but maybe we just didn't try hard enough
Hi there, I'm not sure I completely understand your requirements unfortunately, but have you looked into fine-tuning? It allows you to train a GPT model on data for your own use cases.
https://platform.openai.com/docs/guides/fine-tuning
I was about to play around with it. I am trying to generate some JSON documents that would be valid according to a DSL grammar expressed in JSON. Current models doesn't know anything about this grammar so they don't know the different grammatical rules and how to generate them from scratch.
Does the "Continue Generating" thing use another request? Like 49/50 would go to 48/50?
yes
Yes I have been getting this when trying to use plugins for the past 2 days in Edge. I cleared all the site data, logged back in and still doesn't work.
Hi Steven - I logged out, closed the page, cleared cookies (I believe) and went from there and was able again to access plugins. Seems hit-and-miss but working still.
Hello OpenAI. I have a query about your website. I've uh... been trying to find an answer for a while.
I've reached out via twitter several times, but had no reply. I also tried sending a help request via the chat-message system on the OpenAI website, and... 6.5 months later I finally got a reply in my email that doesn't address my question, and just links me to the FAQ which doesn't answer my question.
So I guess here's my question!
I made an account on OpenAI to experiment with ChatGPT, the only information I provided on account generation was my prefered username and an email address - however when I've been using chatGPT it has a user avatar for me, that's an avatar that I use on a number of other websites.
I didn't provide this data to OpenAI, and the website has no "account management" page where I can change or update it. So... my question is, where did you scrape the image from?
I've asked ChatGPT directly where the avatar comes from, and it says that it doesn't have any information about avatars, it also says that the OpenAI website doesn't provide user avatars (this is incorrect, as I have a user avatar when using chatGPT through the website.)
Clearing the site data should do all of that. But doesn't work. π¦
I'm on MacOS ver 13.2.1 and Safari ver 16.3.
I'm on Windows 11 EdgeDev.
Just starting doing this yesterday.
It has happened to me twice, cannot put a finger on a cause - which means that I have escaped the problem twice too.
I will keep trying, thanks for the help.
How do I use the Prompt Library? Iβm new here and would be happy for any suggestions instead of rolling thru all posts. Thanks π
Anyone else getting unable to load coversation errors for chat history? or is it just me?
Like I do it on PC it's an error but not on the moblie app or if I open up in Chrome on my phone, it's weird.
Good luck. They just nuked most of what made super prompting good. I believe this is why we've all had "conversation history" issues over the last 24-36 hours. I am wondering if they collected super-prompts and nuked them. I had a version of "DAN" but I personally modified it to simply be more human-like and to have a very robust "synergy of personalities" which made for unique responses as opposed to the typical responses that we would normally get. I tried 8 different times in the last hour and it has been shut out cold.
chatgpt uses gravar that links a picture to your email address
right
It would be good if it said that somewhere?
because it sort of looks a lot like it's scraped an image from another account without asking me.
yeah, i originally found it out by looking the page source π€·ββοΈ
ok why am i suddenly hitting this error message? Authorization error accessing plugins
thanks i just uninstalled all the plugins and reinstalled fixed the issue
what bothers me mostly about it is that there's no information about this process on the page - there's no "account management" on the site that lets me set my image there, and when I asked chatGPT where it got the image from, it's like "OpenAI doesn't have user avatars, what are you talking about?"
(oh, and that support couldn't answer the question for me either.)
Hello everyone,
I shared personal info with ChatGPT in an inadvertent moment. I do not wish to let them wander on the Internet in the future. I would like to know if there is a form to request the deletion of this data without having to delete my OpenAI ChatGPT (premium) account. Thank you all
UPDATE: for the person who is in the same situation as me, i have the link of the form
I have but I have not found any solution to this.
Do they share the info with the internet? Idon't think they do, they just use it for data analysis.
the data on ChatGPT may be used in the process of training and improving of the AI unless you opt out of it on the settings menu
Can we generate images on the discord of OpenAI?
nope
ok thx
you can do it here: https://labs.openai.com/
This doesn't mean that your input will be shared. Imagine how easily we could destroy the accuracy of their data by spamming fake information
oh, no, "sharing" never happens in any sense
you are right
but if you don't want to risk it, then opt out
for example some business that don't want to take any chance at any sort of code snippet leaking out
but no conversation history
its also a feature tradeoff
well.. if you don't want OAI to have your chats... they can't have your chats
How do we get the api grants?
Is there no other way to get the grants?
possibly if you are ai researcher with a great cv, startup with an unique business idea etc.
Oh, just the rarest of the rare cases.. Interesting.
Why this? Iβm not use vpn and site worked and app no
Me its inverse... Web site not working but app yes lol
the api is very cheap so there needs to be some justification to actually need a grant, background and purpose is usually always a big factor
Mhm, thanks for the help :)
Hey guys π Am I the only one bored by this new captcha we get after EACH message we send? Annoying as hell i swear
Often it is from user side
- bad addon / extensions from browser
- vpn or vpn proxy
- cookie and cache not properly deleted
hmm, weird stuff, i will try to logout and login again and see if it fixes the issue (hopefully!)
This is a robbery, give me all your chats
me too
ok ok, no need for violence, you can have it
https://chat.openai.com/share/146183ad-3725-4909-9337-528f5986341e

got em
@rigid musk I can confirm i still have the issue after deleting all my cookies, sessions, logout and login again. Lizardcry also have this issue. I'm using Firefox
I swear it's unusable right now. Paying $24/month to check if i'm a human after each messages π
@rain idol@modern minnow the frequency about capchas is directly correlated on how sus the server thinks you are, it is not really a bug, it is working as intended
some things can cause you to get more capchas such as being in the same network as other people who are also using ChatGPT, it is common on schools or with VPN users
another thing that causes more capchas are over intrusive extensions and antivirus who block stuff from running, in this case, it might be simply blocking the methods that check if you are not a bot without capchas, so.. they system resorts to capchas
The only thing which changed from yesterday is the fact i added an extension to automatically solve captcha for another website, but i uninstalled it in the same minute. Nothing changed except this. I'm at home, on the same laptop, same ip, no vpn, nothing like that
And when i say yesterday, it's literally few hours ago. Don't really like the fact i'm paying to get this captcha each time because i'm sus (and for sure i'm not)
if it works on your phone that uses different network then you probably already got flagged on the laptop and might need to wait it out to clean rep
and that "only" thing sounds pretty sus
Do you have a vpn
And if you try with browser on phone it is the samething ?
I do have a vpn but it's turned off anyway. On my phone it's working fine
Make sure web browser and threats protection from vpn are off. They act like proxy and created huge issues
Also can you try with a fresh browser you never use once they are turn down like edge or download opera for the test
But again, few hours ago (15h from now) it worked perfectly fine.
Using Chrome (browser i never used to interact with chatgpt so far) in incognito mode, i still have this issue.
Even with web browser and threats protection disabled ?
I'm using it in incognito mode, so yeah. Only on my mobile it seems to work fine
Incognito doesn't disabled web protection and threats protection from vpn
But again, my vpn is disabled and not used.
Can you change dns in Chrome setting for a test
Sure, how to change my chrome dns and which one should i try?
Should be the 3 dot in top right corner, setting ,security, privacy
You should have a drop-down menu
Found it, what dns should i try?
Can you try the Google one ?
What to do if another person is using my account specially that even if i change the password the website doesn't kick me out to login again
Same issue
I suggest to post on #1070006915414900886 openai do check there and investigate
Why does the site load sooo slow for me tho every other site loads instantly?
its ony chat.openai.com that is slow asf
samee
bunch of people experiencing 502 error rn #openai-chatter
ohh okay
API call failed: module 'openai.api_resources' has no attribute 'FineTuningJob'
anyone else getting this error all of a sudden?
the same API call worked before. it's a regular chat completion, nothing to do with finetuning
update: updating the (python) openai library with pip fixed it
has anyone figured out a system for generating a certain text into a lgo
logo* i mean
I would personally generate the graphical part of the logo first then add in the text yourself with whatever font you seem appropriate (can brainstorm with ChatGPT)
Design Description:
Background & Shape: A circular logo, as circles are inclusive and represent unity - great for the 'All City' concept. The background can be a light shade of recycled gray to keep it neutral and modern.
Cityscape: At the bottom half of the circle, we have a minimalist city skyline. The buildings are various heights but are depicted in simple, straight lines to retain a modern and neat appearance. They should be in a deeper shade of gray, contrasting against the background.
Paintbrush: Instead of a traditional brush, we're imagining a futuristic one. It has a sleek handle and refined bristles. The brush diagonally cuts across the cityscape, as if 'painting' it. From the tip of the brush, there's a streak of green paint which gradually blends into a blue, symbolizing both the planet (green land and blue water) and sustainability.
Company Name: Right above the cityscape, but below the paintbrush, in a semi-circle formation, you have the name βAll City Contractorsβ. The font is modern, sans-serif, and slightly bold. The text can be in deep green to resonate with the sustainability theme.
Optional Elements: You can include small, minimalist leaf designs at the base of the cityscape or near the company name to underscore the sustainability aspect even further.
This design concept is a blend of modernity, functionality, and a touch of futurism. The paintbrush and cityscape interplay emphasizes the company's core business while the color palette and optional leaf elements drive home the sustainable aspect. When realized visually, it should immediately communicate the brand's promise and ethos to potential clients.
this is what i received from chatgpt
Looks good to me! Run that through a text-to-image model and you should be good to go but the name is going to have to be excised from the prompt as text is often rendered incorrectly in such models
do you recommened a certain one
Discussions of non-OpenAI products / models should only be posted and discussed in the #ai-discussions channel.
bruh
π€
how do i contact support, because someone eles has acces to my account
and i dont want to go nothing near the articles, as they are not supportive at all
I have been trying to get a new API key for over 2 days, but each time I get this error message:
"Error generating API key (error=enforcement_failed). If this issue persists please contact us through our help center at https://help.openai.com."
who should I contact or how can I resolve this?
just ask chatgpt how to fixπ
I did lol, gpt told me to contact help center but I can't find any article or answer about this issue
when is GPT-4 finally gonna get the search feature back as well as getting the cutoff at least extended
not known right now, keep an eye on #announcements for updates
Thanks for sharing! Does the prepaid-gated access also apply to older accounts?
nope, not yet
Lol, weird, should I set up a new account then π
no that can cause trouble. at this point you can only wait
"Prepaid billing allows API users to pre-purchase usage. As of today, all new accounts are enrolled in prepaid billing. We're working on opening this billing type up to all users and will also have a feature that allows eligible prepaid billing users to switch to monthly billing."
When will the edit button be available in mobile?
You cannot edit the prompts you sent, instead you can copy or regenerate it
Hi there! Anyone know if chatgpt can be used offline? I mean to use it on a PC without internet connection
nope you cant
Does OpenAI plan to use our conversation history as embeddings at some point so we can RAG through previous history? If they're not going to do this, I'm going to build it
No one else uses chatgpt on my network. It's MY private network thru Google Fiber in Austin. No VPN, as I stated. No proxy, as I stated. Just plain ol DHCP-assigned ip. DNS served up by google fiber isp. Also, disabled all the gd extensions in both Chrome and Firefox.
if something did something to make you look like a bot eg. extension then it might take time to regain the trust level, or maybe it is still interfering
how can I use chatgpt to read data from google sheets? It shouldn't be directly in google sheets, it should only read the data
when is chatgpt 5 coming?
It is not 
It says Code Interpreter, with ability to upload files is enabled, but how do I use this? I'd like to upload a file which is a txt of my LSL code, but don't see how as well as ChatGPT telling me it doesn't know what I'm talking about.
like, they say they aren't aware of any file uploading update and can't see files, but they do know LSL, to clarify.
When you use the code interpreter there will beba icon on the chat box to upload files
Hmm where is it?
At the top of the page where you select the code interpreter under gpt4 section
=P
figured it out thanks
It doesn't appear the feature works properly. When I give my txts to ChatGPT it tells me "I saw a little bit, but the rest was cut off".
Whenever I feed it my 872 line LSL script in TXT format, it cuts most of it off, and when I give 200 lines from my raycasting script with priorities, it cuts off everything beyond about 100 lines in.
am I allowed to use content I generate with ChatGPT in commerical applications?
sending files to code interpreter is not a bypass on the context limit, there is no such a thing
the files you send are supposed to either be code to run or data to be fed to the code to run
the AI only looks at the first few lines of the files you provide to have an idea of what kind of data it is
"since I can't type this 800 line file into the chat, I will upload it to code interpreter" β
"write a script that sums each line of this .csv file and give me the output as another csv file" β
How do other gpt apps have gpt-4 before the original chatgpt app
oh, I thought it was where you upload your code to the AI and it can read the code file so it can help you easier given the too-short limit otherwise where I keep having to split my messages into 3s or 4s.
it might be good to clarify it's not actually for your code files
the file upload is for the code interpreter to be able to run code on your files
not for the AI to read more text at once
The same happens with the link reader plugin though, I send the links to the AI and it tells me it got cut off shortly into reading the webpages. It happens with Pastebin, Wikipedia, everything
yea, that confuses people very often
i have a huge prompt convo going on, and its slowing down my browser. is that normal?
are you on CHatGPT or playground?
actually, this can happen anyways, yes, it is normal
happens a lot on the playground
and less often on ChatGPT
Hi guys, my new credit card got blocked by openai again, after one month's Plus subscription
Does anyone have a solution for this?
OpenAI's support never responded
I'm in Singapore and my cards are absolutely legal
How do I access the Open AI Plugins for Chat GPT? I am a plus subscriber but do not see any option to utilize the plugins.
Settings -> Beta Features
Has the openai 5 dollars free trial been disabled or something? I heard there was and created my first account but there's nothing
is the site down or what? it doesnt load
It expired after some time
So no more free trials?
If not used no, you can always discuss with support@openai.com
Why does my account have no free in it then?
(also I couldn't find the support section in openai, mb)
just add a payment method and then you can use the API
Hello
Can I use openai to create a model in js to create a chatbot or do I need to use python for backend and js for frontend
you can use the API on anything that is able to do a http request
there are libs for almost every language https://platform.openai.com/docs/libraries/community-libraries
including python and JS
thanks
Hi, today i can't use chatgpt+ anymore. I've been a subscriber for over a month but now I'm being asked to subscribe. I sent a message and screenshots on openai messaging but I have no response. Does anyone have the same problem as me.
what prompt to ask chatgpt to produce inline table rather than ascii table?
so i prefer this kind of table
rather than this kind of table
can i pay for the plus subscription with paypal?
Credit card only for now .
okay thank you
"markdown table outside of a code block" maybe
Hi there, Im a Chatgpt+ subscriber and have gotten the "something went wrong" error for 4 days in a row now. Any possible reasons/solutions for this. Thank you!
Where's the code interpreter on plus?
I'm currrently on Plus plan and can't see it
You need enable in setting
No problem, have a good day :3
both are the same, the AI always output an ascii table, but the page displays it as a embeded table
it just happened that in this case, it decided to put it on a code block
hi! New to discord.. trying to get some answers - maybe they're easy for someone on here and i just overlooked it !
I have an openAI account -- i generated three keys thus far, newest one today -- but when i tested it to get today's date -- it said August 11, 2020.
does anyone know why?
api does not know the date unless you tell it to it
another plug in i used with openAI was "today's date is October 25, 2022."
interaxAI plug in (first one was mind2matter) ...
anything special or outdated with these dates that i didn't "update?" lol
what are you doing or using?
hi @trail cairn ! just generated a key from openAI and plugged it into mind2matter plug in for my website and am using another plug in interaxAI with same key
honestly just using and testing questions through the plug ins
you need to contact authors of those products if you are just using an api key with them
@trail cairn -- so i'm in here : (picture attached) https://platform.openai.com/playground and the date comes out wrong again
yeah but api does not know the date if you dont tell it
gpt = pretrained
So enterprises will get GPT-4 with 32k context and uncapped usage? Really?
Would love to see the prices.
how to get discord auth code?
Hello I'd like to enquire about ChatGPT Enterprise
will openAI release a paid tier between chatgptplus and chatgpt enterprise for the common user
Kinda self explanatory
not exactly, how is it self explanatory?
Enterprise is for enterprise, plus is for normal people
so, will there be a better tier at a higher price for normal people?
so chat can do 32k context now - are messages still less than that? e.g. my API still not approved for 4-32k but have a couple of things I'd like to route to the 32k model
any good way to do it through chat?
why?
but programmatically
Why not just use the api?
using the sessionID
if talking abt chatgpt 4, far more expensive,
cause no
package that will route to chat and then pull from it using sessionid?
something similar, but yes
So you want Enterprise levels of function without paying for them? I don't think I can help with that a whole lot
not really, an individual does not need enterprise level of function, the only benefit could always be removal of the message limit?
I'd like to pay for it, been waiting for 4 months for the 32k api and spend up to 1k/month some months but no access yet
If you don't like the message limit, why not use the api? If it's too pricey, there's always ChatGPT Plus
Source?
Trust me
So let's not spread misinformation π
Imagine if they just stopped developing gpt, no
so any other ideas for how to get 4-32k other than plead every few weeks
Please keep the ChatGPT outputs in #1050184247920562316
I don't think so, unfortunately. I, too, have been waiting quite a few months for it π
bummer
When can regular subscribers expect to tap into the 32k context size, especially without the per-message cost? Are there any plans or roadmaps that shed light on this?
Elevated security might be important for businesses seeking a competitive edge. They don't want their assistant helping rivals after learning how to help THEM through plenty of iterations. But at the same time, training data is OPENAI's leverage. It refines the AI. Elevated security deprives OpenAI of training data, making it a cost factor, especially with other AI's becoming more capable. The pricing strategy becomes a balance between ensuring data acquisition and offsetting potential losses from enhanced security measures.
I am fine with the idea that my interactions could further enhance ChatGPT's capabilities.
Appreciate any feedback or updates on this. Thanks! (spellcheck edit)
I don't think there's a road map as to when the 32k model will be available to the general public (if it even happens). We'll have to wait for a message in #announcements about it π
How to get more DALL.E credits?
you can use the /daily command in #daily-theme for 4 free credits a day, or you can purchase 115 credits for $15 USD (plus tax) on the labs website
Forget the 32k model they havenβt even given us the 16k model for 3.5 as ChatGPT plus members , only through API so I wouldnβt hold my breath.
I wouldn't expect them to offer the 16k 3.5 model on ChatGPT's website. I just want API access to them π₯²
how ai is made ?
You can try the "'Something went wrong.' instructions at https://help.openai.com/en/articles/7996670-chatgpt-says-something-went-wrong? Maybe refresh the page or close it and reopen it. Try signing out and signing back in. Clear your browserβs cache?
Great question. Since you are in OpenAI's Discord channel, I can refer you to OpenAI Research method that leads to AGI at https://openai.com/research/overview. (A POV: we are using AI to make AI. There is a manual approach vs. AI generating algorithms to genrating AI.) #ai-discussions
is gpt3.5 16k available to new users that just signup?
Check @bronze charm's response in #openai-chatter . You should be able to enter gpt-3.5-turbo-16k into your API settings.
The GP be T-ing
Getting the ole "something went wrong" no matter how I prompt, anyone else?
I followed the something went wrong wiki. Cleared cache, made sure VPNs were disabled. Tried enabling a VPN. Signed out, signed in. Incognito window. Nadda
Works in firefox but not chrome....? Weird
Some users have reported the 'Something went wrong' error in Chrome but not in Edge, on the same network. Maybe a plug-in you are using? The ChatGPT error messages FAQ https://help.openai.com/en/articles/7996703-chatgpt-error-messages recommends a different network and turning off URL blockers or extensions, besudes disabling any VPNs.
what is wrong with it. its been doing this recently
Yes, this just started for me Friday evening, turned off the VPN, flushed the website cache, hard reboot, switched to 3.5, back to 4, back to 4 without Code, tried every day, and nothing. -My android does still work though. (I tried on Edge, and it did not work either, same error (something went wrong...)
--Any news on how to address this?
*I turned off all my extensions except some google docs ones
Were you able to resolve this? Mine stopped working as well, with little to no reason
Thank you for the update!
The 32k model being offered to those from top-tier business environments (S&P 500 professional types) implies that the model meets a quality standard that is reliable and that certain technical hurdles have been accounted for.
I'm excited about the broader potential if such advanced capabilities were to be more widely accessible. While it's reasonable for major businesses to have initial access, I'm hopeful that individual enthusiasts and innovators won't be left waiting too long.
I've also observed the name transition from "Code Interpreter" to "Advanced Data Analysis." This transition in terminology is a bit intriguing, as the term "Advanced" carries some weight. It seems to hint at functionalities beyond traditional data analysis, perhaps even at the potential of real-time, web-based operations, especially given that even foundational data-analysis platforms, tools like Microsoft Excel, have integrated some form of real-time online analysis capabilities.
Considering the research indicating ChatGPT's marked improvement in mathematical performanceβjumping from 50% to over 90% in tests where even top human scores hover around 96%, and this improvement being reliant upon this "code interpreter" or "advanced data analysis" aspect, it would be nice to see how this tool develops. Will ChatGPT eventually be able to use plugins with the advanced data analysis capability activated in one conversation tab? (not just separately?)
OpenAI's ongoing advancements with ChatGPT are genuinely commendable to witness. I'll be following future announcements with keen interest and vigilance.
I don't work for or represent OpenAI, so I can't answer your question unfortunately π«€
Sounds like a confusion.
Avast webshield can cause this. Disable that. Or other web protection solutions can cause similar issue.
in how many years do you guys think that 3.5 will be free
Yo guys, i saw this new chatgpt enterprise thingy, how do i know if i alr got it? I see no diff in my interface other than the code interpreter being renamed
Anyone have a clue how long it takes for OpenAI to raise the api key spend limit? I've been stuck at $120 spend now for 4 days and it's really annoying since I need this key for everything
Guys. is there a Limit of Messages on ChatGPT
Ty gpt
where do i report chatgpt errors?
yes, both 3.5 and 4 have limits
you should have planned ahead, it should normally take a week if approved but seen people also say weeks
I mean this is the first time a company has forced a stop on my spending for usage lol I didn't know the $120 was something you couldn't just raise manually yourself
pretty unfortunate
I am not a big corporation but still I have access
it just means you got lucky at one point when they still had capacity and did random draws, since then not even some people who spend thousands to api monthly have received access
Consider using anthropic version!
when potential new model news
new dalle 3 or
gpt-5 or
free chatgpt(gpt-4) or
gpt-4 unlocked image reading
2024-2025
even MacOs chatgpt app that have shortcuts support would be great
would be great if they make api shortcut functions on it too rather than just chatgpt
im not sure if you are asking how you got it, or just flexing π
How did u get it
When I try to use GPT 4 API, I get this error:
Error with OpenAI API request: The model
gpt-4does not exist or you do not have access to it. Learn more: https://help.openai.com/en/articles/7102672-how-can-i-access-gpt-4
So I don't have access to the GPT 4 API, even though I'm a paying customer and I had registered for the waitlist many months ago!
OpenAI's press release said that anyone who paid for GPT usage before August gets access to GPT 4, so why didn't I get access?
Same error in the playground btw
I was using chatgpt-4 for usual dev questions and suddenly got an error "We have detected unusual activity from your system... please try again" then it all stopped working and i can't use anything anymore. Any ideas?
Is chatGPT down?
Oh, I see Partial Outage on the status page?
hello guys , oI just have beeen blocked for no reason. Anyone else ?
Welcome to OpenAI's home for real-time and historical data on system performance.
Thx you @charred parcel
ye it's down, got blocked on chromium to verify here, but not on firefox, random.
i can't generate stuff anymore.
chatgpt's GUI is down
which means it doesn't work.
Seems like service is down and they are investigating
and they're asking me to subscribe again
oh haha went through the exact same thing in that order lmao
how long will it be down? I need it for work
I guess we just have to wait
i firstly thought it was my vpn, but that's just blocking google from me logging on.
but it's definitly an issue they are investing in, the time is unknown.
Same error hereβ¦ It also show the advice βupgrade to PLUSββ¦ so I thought I had to renew my plus subscription but it doesnβt process the payment..
Same here.. I thought that my subscription has to be renewed⦠But anyway it cannot process the payment..
hi guys... im new here.. is there a problem with chatgpt today? im not a plus account holder.. just wondering if its only happening for free accounts?
Is chatgpt currently dead?
Same ..here and I have a plus account...very strange
yes it's down even for plus accounts
and my account says you have been blocked..
I'm a Plus Account holder, I also started getting issues 15 minutes ago. I think this might be world wide. My account has also been blocked
I'm sure they will fix it soon
i was on an German vpn that's hosted on my vps, and i thought that the network was down or that my vpn was blocked, which is just hosted myself lol
but it is just an incident at OpenAi.
Bro i had a project to do, due tmmrw and chat gpt not working rn
eu user here, can confirm problems
me too
same π₯²
Go for a run or something. I'm sure it will be back before we know it
was working with ChatGPT and then all my history disapeared and then it would not load forever and when it did, no more history and it was saying i needed to upgrade to Plus version which I am paying for....and now it blocked me completely with a Cloudflare Ray ID: 7fe57ff76c724c16
Same here
I don't know why my account was banned
seems like its down yes
Welcome to OpenAI's home for real-time and historical data on system performance.
π
good channel
pls fix fast
hey i was blocked i dont know what "attack" means when i used it like always
Hmmmm
was i banned?
I don't know why my account was banned
when i was logged in it would not show either my conversations nor would generate any output or error at all
i logged off and tried to log in this happens i have neither received any email from open ai either
i was able to use it like an hour ago
I am having the same issue as paid subscriber
strange it showed me as another account above lol
everyone can use the enterprise plan with the plus account?
ahh nno it was another person who got blocked as well