It was really helpful. Thank you very much! Just one think. The code as it is allowed me to send only one message. I had to modify a line as below. [before] [after] handleSend(message)} /> The credit is to ChatGPT who found the issue and suggested the new line. As a person with 0 experience in react or even JavaScript, this was daunting to me. I see now that it is still very important to know your stuff, otherwise even with AGI, you will not be able to ask the right questions.
A friendly reminder for folks who are trying this out - at 12:46, code Line 52, remember to add a space after your word "Bearer ". Otherwise the API calls will fail for not being able to find the API key.
@@kanizfatema755 That means you've either used up your free tokens or your free tokens have expired, the tokens expire after 3 months have passed since you've signed up with that specific account.
Thanks so much! It works great, and following your video did not take much time . Presentation-wise it is a masterpiece. Using this as a template, I will repeatedly refer to your channel.
The API used in this video is different from the new Assistants API, right? Could you quickly explain how the code would differ using that new API, or is it that more of a topic for another video?
Does this code has the ability to continue a conversation, that is, can remember the previous responses and prompts while giving any new response like that of a chat in ChatGPT?
it's not worked for me , this is the error : Failed to load resource: the server responded with a status of 429 () Uncaught (in promise) TypeError: Cannot read properties of undefined (reading '0') at App.jsx:79:23 at async processMessageToChatGPT (App.jsx:66:5) at async handleSend (App.jsx:36:5)
Has anyone else encountered issues with all messages being aligned to the right side? I've compared my code to Cooper's and it looks almost identical so I'm not sure what the problem is... Many thanks in advance
If I want to train the bot on specific information about my business or website, do I put this information in the SYSTEM role? So that when customers on my app ask the bot it can respond to them with specific information I've provided? And does this system data contribute to token consumption?
I actually need to look into the specifics of this, as I believe it does depend on the system prompt. I will release a video discussing the system prompt in specific and how it changes the ChatGPT API soon 👍
I am also searching this type of video..I also want to train the bot for specific information..if you found out please reply to this comment...it will be very thankful..😊
Two questions. But first, thank you. I look forward to following you. A huge improvement over what I had been using. My questions are two simple formatting issues. First is there a simple code input to have the text fields left-justify instead of center-justify? [Edit: I found justification of text can be accomplished in the accompanying "App.css" file in the first section "#root" under "text-align:". Other UI formatting can be done in the "index.css", such as colors, etc.] And second, the scrolling seems wonky after the chat history fills the chat container... is there a fix for this or are those issues only addressed through the ChatScope UI? Thanks again for your time and sharing your knowledge. Much appreciated!
Why my code show error like Uncaught (in promise) TypeError: Cannot read properties of undefined (reading '0') at App.jsx:79:23 at async processMessageToChatGPT (App.jsx:66:5) at async handleSend (App.jsx:36:5) Are you know how to solved it?
@@daftgin yes. The problem is from the trial was over. So you must subscribe it, and will be work. Notes: you cant use with make a new account just to get a free trial again if u used a same telp number
thank u! i got it working local on a mac which is amazing. i wonder how do i use it on my webserver.. i ive triied the same process via ssh but i get issues .. clearly i dont do npm run dev .. thanks again!
Nice Video. I have a proplem somehow...The first response from ChatGPT is always some random stuff. After the second fetch, i get the answer for my previous question...What am i doing wrong?
Im having trouble on 6:15. Suddenly he skips something... The [vite] hmr update in the Terminal is not shown in my Terminal. Also, when I do want to open the project in browser it says Edit src/App.jsx and save to test HMR. What should I do? Any help or hint is appreciated!
can you make a video to show us how to turn it into a simple to use android app, we don't always carry the computer with us, while portable devices next to me are very common variable, and especially you should set the maximum number of characters it can handle, and add a minimum response time of 30 seconds to get the most complete response, we look forward to hearing from you. ,
@CooperCodes I ran the app in my local system with this command (npm run dev), but it is stuck and keeps saying that chatgpt is typing. Am I missing anything?
I have a problem at min 6:00. Once I want to open the chatbot in browser, it is telling me that I should Edit src/App.jsx and save it to test HMR. So what did i do wrong? (btw I fully downloaded the code in first place)
This is awesome. I downloaded and ran your code. Just curious on one thing... why does it seem so slow to complete the responses? I'm seeing just under 10 seconds per response.
That unfortunately has to do with the speed in which the ChatGPT API can process requests. I think a BUNCH of people are using the API right now and so they are having to scale up their services to meet the demand. Hopefully in the future we can count on faster responses from the API, but hopefully this clears things up.
The UI shown in this video was made for example purposes, but its highly applicable to any case as the saving of the messages is the most important part. You can make a custom UI that works similarly (shows messages in a stack like in the video) or you can modify the @chatscope/chat-ui-kit-react library which has documentation here: chatscope.io/storybook/react/?path=/story/documentation-introduction--page . This is a complicated answer but hopefully helps!
Hi, I am getting an error message as "Too Many Requests with status code: 429" and I am getting as reposne "Error communicating with OpenAI API" Can anyone help me with this?
It seems to, not only remember the previous prompts, but is resubmitting all of them each time you try a follow up prompt. Does anyone have a solution to this issue? This is an amazing tutorial, thank you so much.
Great question, use one of the solutions provided here stackoverflow.com/questions/55830414/how-to-read-text-file-in-react . Getting text from a local file isn't as intuitive as you could hope in React, but eventually with one of the solutions there you will get a string from your text file. You can then set the content of your system message to the string, like content: textFileContents . Hopefully this helps!
I'm getting this error - You exceeded your current quota, please check your plan and billing details. Although I have 18$ credit that I haven't used at all
You get $18 worth of tokens for free whenever you create a new account, and $18 goes a really long way for this API when it comes to testing and even small production environments. All the calls from this video cost around 1 cents I believe, so you get enough tokens to test it out / see if it works for your use case. Hopefully that helps :)
@@CooperCodes Thank you for responding. Your code is very useful. I am currently creating a virtual assistant for my company. I am trying to make the code recognize when it is asked for specific information so that it does not make it up. We have that type of information in a data spreadsheet, and my plan is to tell the API something like: "Answer this question, but use the information from this spreadsheet for your response" (Obviously, only when the code finds that it is being asked for specific data from the spreadsheet. This way, we avoid the problem of it providing false information about specific topics it doesn't know). I learned a lot from your video and your repository. You have earned another subscriber. Greetings from Chile :)
when I call api request it says in response in network tab : "You exceeded your current quota, please check your plan and billing details." "insufficient_quota" PLEASE SEE this and fix this!
I have a Discord bot running with 'text-davinci-003' and am too noobish to figure out how to properly convert this to a 'GPT-3.5-Turbo' ran bot instead. Are there some easy fixes to this, like a simple checklist to go through and be sure everything's good to go?
No worries I got you. The API call in here should be applicable to SvelteKit (but managing the messages is gonna be a little different with the svelte stores). Thanks for watching btw!
You would have to make custom components, I used an external library to make things easier but it is not required. If you want to do things yourself I recommend following this line of thinking: 1. Create a chat message component , 2. Create a way to get access to all the chat messages (so you can store them in the API call like this video). Hopefully this helps.
![ทุกครั้งที่หลับตา (Lucid Dream) - AYLA's [ Official MV ]](http://i.ytimg.com/vi/4i1OBAQyv58/mqdefault.jpg)
It was really helpful. Thank you very much! Just one think. The code as it is allowed me to send only one message. I had to modify a line as below.
[before]
[after]
handleSend(message)} />
The credit is to ChatGPT who found the issue and suggested the new line. As a person with 0 experience in react or even JavaScript, this was daunting to me. I see now that it is still very important to know your stuff, otherwise even with AGI, you will not be able to ask the right questions.
A friendly reminder for folks who are trying this out - at 12:46, code Line 52, remember to add a space after your word "Bearer ". Otherwise the API calls will fail for not being able to find the API key.
yea that's why I hate the old "" + notation, just use a string template literal
Brother I'm getting error in this line how to solve
Hey I am getting this error you execced your current quota please check your plan can anyone help ?? Why I am getting this error??
@@kanizfatema755 That means you've either used up your free tokens or your free tokens have expired, the tokens expire after 3 months have passed since you've signed up with that specific account.
@@kanizfatema755 did you solved?
Awesome work. Idea for a part 2: add the hability to speak to the chat with whisper and to have and audible response by the chat.
thats a really interesting idea
This tutorial is for who thinks on a new Idea with openAI API good work bro
Thanks so much! It works great, and following your video did not take much time . Presentation-wise it is a masterpiece. Using this as a template, I will repeatedly refer to your channel.
Amazing tutorial. And I love that you're using Vite
Thank you so much, your presentation is excellent!!!
The API used in this video is different from the new Assistants API, right? Could you quickly explain how the code would differ using that new API, or is it that more of a topic for another video?
Amazing tutorial. And I love that you're using Vite
I will make a best friend for myself now thank you!
You are just a fabulous person, sharing great informations.
is this api need to be paid ?
@@PrabuLapxx few hit tokens are free. But major are charging
Amazing! Thank you so much!
also add pinecone and their algorithm example like semantic search video audio search using open ai and pinecone
thank you so much! helped me a lot
Nice content here!
Thanks for sharing
Does this code has the ability to continue a conversation, that is, can remember the previous responses and prompts while giving any new response like that of a chat in ChatGPT?
it's not worked for me , this is the error :
Failed to load resource: the server responded with a status of 429 ()
Uncaught (in promise) TypeError: Cannot read properties of undefined (reading '0')
at App.jsx:79:23
at async processMessageToChatGPT (App.jsx:66:5)
at async handleSend (App.jsx:36:5)
Rich in tutorial content. Thank you.
Thanks this was very helpful
@CooperCodes Awesome tutorial, thanks so much. Do you know why we have to send the full array of messages every time to the API?
Awesome tutorial. Try to explain about all the stuffs a little bit deeper . But love it
Thx for this great tutorial!
Awesome video!!!
Thank u very much!!!
Excellent! Thanks for posting. Do you have a code repository? Would like to try to learn from your example, test, fail and test again😂
Yup! It is in the description, would just copy here but I’m on mobile 💀
It would be great to see you, build something using the stream API
Excellent! Totally workable
Is it working
@@uparagnath5409yes of course!
This is AWESOME! Thanks Cooper!
Do you know how to get the "best_of" parameter working?
Has anyone else encountered issues with all messages being aligned to the right side? I've compared my code to Cooper's and it looks almost identical so I'm not sure what the problem is... Many thanks in advance
Update: it turns out I needed to specify on all messages the direction, either "incoming" or "outgoing" fixed the issue.
this is a really, really good video
Which chatgpt model are you using?
@@shrutinag889 3.5
Hello, I've got problem with CORS policy, andy idea how to fix it?
If I want to train the bot on specific information about my business or website, do I put this information in the SYSTEM role? So that when customers on my app ask the bot it can respond to them with specific information I've provided? And does this system data contribute to token consumption?
I actually need to look into the specifics of this, as I believe it does depend on the system prompt. I will release a video discussing the system prompt in specific and how it changes the ChatGPT API soon 👍
@@CooperCodes he may want to check into embeddings. Langchain will allow this
This would be really interesting to know as i was looking to do the same thing.
@@CooperCodes hey cooper have you looked into this problem ??
I am also searching this type of video..I also want to train the bot for specific information..if you found out please reply to this comment...it will be very thankful..😊
Two questions. But first, thank you. I look forward to following you. A huge improvement over what I had been using. My questions are two simple formatting issues. First is there a simple code input to have the text fields left-justify instead of center-justify? [Edit: I found justification of text can be accomplished in the accompanying "App.css" file in the first section "#root" under "text-align:". Other UI formatting can be done in the "index.css", such as colors, etc.] And second, the scrolling seems wonky after the chat history fills the chat container... is there a fix for this or are those issues only addressed through the ChatScope UI? Thanks again for your time and sharing your knowledge. Much appreciated!
Why my code show error like
Uncaught (in promise) TypeError: Cannot read properties of undefined (reading '0')
at App.jsx:79:23
at async processMessageToChatGPT (App.jsx:66:5)
at async handleSend (App.jsx:36:5)
Are you know how to solved it?
Have you solved this?
@@daftgin yes. The problem is from the trial was over. So you must subscribe it, and will be work.
Notes: you cant use with make a new account just to get a free trial again if u used a same telp number
@@verrr23 I was able to fix it but my situation is different, I just had a typo in my code
great, thanks man...any way to make the chat responsive?
i tried to build the app same as your code but there is an error in console: (429) (Too Many Requests)
Same here! Did you solve it ?
nice and clean
Is there a way to use with a specific GPT created on MyGPTs section?...
Thank you!
Hi
Firstly thank you so much for bringing this tutorial with reactjs only. just one query though, My response is very slow can help me with that ?
Great Video ❤
Thank you so much!
you are great man!!!
With this library, can I edit the styling of chatbot for example adding a logo/icon before the chatbox or changing the background colour?
thank u! i got it working local on a mac which is amazing. i wonder how do i use it on my webserver.. i ive triied the same process via ssh but i get issues .. clearly i dont do npm run dev .. thanks again!
How could we add Llama index to this project to have a custom knowledge base?
Nice Video. I have a proplem somehow...The first response from ChatGPT is always some random stuff. After the second fetch, i get the answer for my previous question...What am i doing wrong?
can i style the component like i would to change the color of MessageInput box?
hello I followed the tutorial to the letter unfortunately in my terminal I have error 429
, how to solve it?
I have the same problem, did you solve it .
amazing i did it!)
Thank you very much
Im having trouble on 6:15. Suddenly he skips something... The [vite] hmr update in the Terminal is not shown in my Terminal. Also, when I do want to open the project in browser it says Edit src/App.jsx and save to test HMR. What should I do? Any help or hint is appreciated!
how we increase the accuracy of an api key get an immediate response?
can you make a video to show us how to turn it into a simple to use android app, we don't always carry the computer with us, while portable devices next to me are very common variable, and especially you should set the maximum number of characters it can handle, and add a minimum response time of 30 seconds to get the most complete response, we look forward to hearing from you. ,
Whenever I type anything to it, it just says chat-gpt is typing and doesn't provide a response. Do you know why?
Same error....did you solve it??
@CooperCodes I ran the app in my local system with this command (npm run dev), but it is stuck and keeps saying that chatgpt is typing. Am I missing anything?
I have the same problem, did you solve it .
Thank you
I have a problem at min 6:00. Once I want to open the chatbot in browser, it is telling me that I should Edit src/App.jsx and save it to test HMR. So what did i do wrong? (btw I fully downloaded the code in first place)
Why i can't upload my react work with OpenAI API to Github repository?
This is awesome. I downloaded and ran your code. Just curious on one thing... why does it seem so slow to complete the responses? I'm seeing just under 10 seconds per response.
That unfortunately has to do with the speed in which the ChatGPT API can process requests. I think a BUNCH of people are using the API right now and so they are having to scale up their services to meet the demand. Hopefully in the future we can count on faster responses from the API, but hopefully this clears things up.
Hey I am getting this error you execced your current quota please check your plan can anyone help ?? Why I am getting this error??
Hi, this is because openAI has limited a free access to APIs to around 2-3 months. You can use the key from the newly created account.
Amazing stuff! How can I style those components?
The UI shown in this video was made for example purposes, but its highly applicable to any case as the saving of the messages is the most important part. You can make a custom UI that works similarly (shows messages in a stack like in the video) or you can modify the @chatscope/chat-ui-kit-react library which has documentation here: chatscope.io/storybook/react/?path=/story/documentation-introduction--page . This is a complicated answer but hopefully helps!
Can you deploy this using Streamlit?
Hi, I am getting an error message as "Too Many Requests with status code: 429" and I am getting as reposne "Error communicating with OpenAI API" Can anyone help me with this?
why does my chatbot just freeze on typing
It is possible to build on python? (i have 0 knowledge with coding language)
Does the app preserve the responses?? Like if I refresh it will it still remember the prev responses?
It seems to, not only remember the previous prompts, but is resubmitting all of them each time you try a follow up prompt. Does anyone have a solution to this issue? This is an amazing tutorial, thank you so much.
no, in chatgpt it does the same, every time you make a new prompt it also send the previous
This was a great tutorial! However I do have one question, how can I make it so that my system messages content is derived from a text file?
Great question, use one of the solutions provided here stackoverflow.com/questions/55830414/how-to-read-text-file-in-react . Getting text from a local file isn't as intuitive as you could hope in React, but eventually with one of the solutions there you will get a string from your text file. You can then set the content of your system message to the string, like content: textFileContents . Hopefully this helps!
i followed step by step but the chatgpt didn't work well with me
This will work on wix website?
Why doesn't typinginput appear when deployed?
I'm getting this error - You exceeded your current quota, please check your plan and billing details. Although I have 18$ credit that I haven't used at all
yes same error im getting, whole project wasted
@@pranav5532 it kinda worked... Try to change the model to gpt-3.5-turbo and hopefully this will work with you too.
@@emmyyassin1587 Am getting the same error too , any solutions ? am using 3.5 turbo only
I am using gpt-3.5-turbo and still getting the same error ?? Can anyone help??
@@emmyyassin1587no this doesn't working do you have any other solutions??
This codes works on wix javascript velo code?
is it possbile to connect supabase? to custom chatgpt messagner
hello guys is the api free for all or i must buy one ?
not working in my case
what should i do
Does anyone else get the TypeError: Cannot read properties of null (reading 'useRef') error?
Hi, I can't register on the site (because of the embargo). Is there another way to get the OpenAI secret key? Or can I use the key in the video?
why is it not working for me? even copy pasting your code isn't working.
no bullshit just perfect
How do I program a personality or a historical character in the CHAT GPT?
you would have to put it all in the system role, or find a way to summarize user chat history and put it in there.
Great 🔥
It’s not working in windows 11 can any one explain
it is not working for me can anyone help,the bot keep saying typing please help
very useful
Glad you found it useful, thanks so much for watching!
Why is my chat option not working and showing error?
Why nobody is explaining how to train LLM model with our own custom data in react..
Feel free to take a look at my more recent video using Supabase and OpenAI for that.
I'm getting this error - You exceeded your current quota, please check your plan and billing details.
Any solution?
i have this same error too,
Some help please.
same
x2
You need to start paying for OpenAI credits (you ran out), and then double check you are using the API key for the account you paid for.
It doesn't respond to entered questions what should I do
I am geeting in post method api fetching process?? Any one solve this
Chatgpt verson issue??
I don't understand, I thought we had to pay to use this API. How are we using it?
You get $18 worth of tokens for free whenever you create a new account, and $18 goes a really long way for this API when it comes to testing and even small production environments. All the calls from this video cost around 1 cents I believe, so you get enough tokens to test it out / see if it works for your use case. Hopefully that helps :)
@@CooperCodes Thank you for responding. Your code is very useful. I am currently creating a virtual assistant for my company. I am trying to make the code recognize when it is asked for specific information so that it does not make it up. We have that type of information in a data spreadsheet, and my plan is to tell the API something like: "Answer this question, but use the information from this spreadsheet for your response" (Obviously, only when the code finds that it is being asked for specific data from the spreadsheet. This way, we avoid the problem of it providing false information about specific topics it doesn't know). I learned a lot from your video and your repository. You have earned another subscriber. Greetings from Chile :)
Hi, I am getting an error message as below
when I call api request it says in response in network tab :
"You exceeded your current quota, please check your plan and billing details."
"insufficient_quota"
PLEASE SEE this and fix this!
You can say it requires pricing in tokens , like 1K tokens is 750 words , am i right? i think i should pay first, then it will run , right ???
Hi, this is because openAI has limited a free access to APIs to around 2-3 months. You can use the key from the newly created account.
this APP DONT WORK ERROR TYPING [0]
when I ask for code, i cant see the code, is there a way to implement a code block in the response?
I have a Discord bot running with 'text-davinci-003' and am too noobish to figure out how to properly convert this to a 'GPT-3.5-Turbo' ran bot instead.
Are there some easy fixes to this, like a simple checklist to go through and be sure everything's good to go?
I tought it was going to be svelte :(
Aye sorry, will do an AI powered application with SvelteKit soon!
@@CooperCodes Thanks for this tutorial tho!
No worries I got you. The API call in here should be applicable to SvelteKit (but managing the messages is gonna be a little different with the svelte stores). Thanks for watching btw!
be aware - you left the api key in your github commit
Yup I disabled it, I like to keep it there so people know what to replace. Thanks for this tho :)
is this api need to be paid ?
como modifico todo con tailwind?
You would have to make custom components, I used an external library to make things easier but it is not required. If you want to do things yourself I recommend following this line of thinking: 1. Create a chat message component , 2. Create a way to get access to all the chat messages (so you can store them in the API call like this video). Hopefully this helps.