The problem with beta assistant api is that it is a token hog. If you create a thread, the api stores and returns all conversations up to the limit of tokens supported by the model, and does not give the option to remove conversations from the thread. And the solution of creating new threads every time doesn't seem smart.
I have 2 questions: 1: Do videos remain embedded in the chat window after you send next message or refresh the page? 2: Could you provided the json for the function on open ai? It would be great if you could answer those 2. Thanks :)
Hi @kie Codes, Do you have time to help me with this video for the assistant ID? I'm having trouble, and here is the error I'm encountering: Unrecognized request argument supplied: assistant_id , invalid_request_error (truncated...) somebody help me please !!!
I really want to get this running locally on my machine. But when I do npm install, I get a moderate error when running this code that says “PostCSS line return parsing error”. Do you know how to resolve this issue for this code ?
@@KieCodes For example I have integrated asissistant to bot press . but asisstatn itself has no gpt4 vision function and no audio to text transcription . Can I add assistant api + gpt 4 vision and transribation to bot press to create a chat bot for whats app and instagram.
Yes you need to add functions for the different database queries you want to do. Like get_user_account_data(username) and this will then handeld by you backend software and be handed over to the assistant to give a proper answer to the user.
Thank you. I started playing with Assistants API and this video helped me a lot.
I am glad it helped! This motivates me the most! Thank you! 🙏
Please rebuild this with a node back end... Ill buy you a coffee :D
Good idea!
@@KieCodes I second this, id love to see a nodejs version of it, can't find one anywhere.
The problem with beta assistant api is that it is a token hog. If you create a thread, the api stores and returns all conversations up to the limit of tokens supported by the model, and does not give the option to remove conversations from the thread. And the solution of creating new threads every time doesn't seem smart.
great video, help me a lot
I am glad to hear that!
I have 2 questions:
1: Do videos remain embedded in the chat window after you send next message or refresh the page?
2: Could you provided the json for the function on open ai?
It would be great if you could answer those 2. Thanks :)
amore details - excellent - thanks
Thank you!
Hi @kie Codes,
Do you have time to help me with this video for the assistant ID? I'm having trouble, and here is the error I'm encountering: Unrecognized request argument supplied: assistant_id , invalid_request_error (truncated...) somebody help me please !!!
I really want to get this running locally on my machine. But when I do npm install, I get a moderate error when running this code that says “PostCSS line return parsing error”. Do you know how to resolve this issue for this code ?
Hey there. Never had this. Maybe join my Discord and I or one of the other coders can help you.
Long time no see!
Yes mate. Glad you’re still around.
thankyou for the video.Kindly share do i have to create sepeate api/route folder for sending request from api.ts api/** in in next js 13.
Hey Kie, awesome code man, is the build supposed to be missing? Thank you btw for this code, but I cant seem to find the nodes and build?
Hey there, the link to the GitHub repository is in the video description. Cheers
Hello, thanks for the video. Is it possible to add GPT 4 vision and voice transributions to the assistant api?
You are welcome! 🙏 Yes you can use all the different APIs in one software.
@@KieCodes For example I have integrated asissistant to bot press . but asisstatn itself has no gpt4 vision function and no audio to text transcription . Can I add assistant api + gpt 4 vision and transribation to bot press to create a chat bot for whats app and instagram.
You can use assistant with function calling.@@Nursultan_karazhigit
Great video, thanks. Do you take on private clients to assist with implementing the above?
You‘re welcome. Sure. For business inquiries please use: stdin@kie.codes
Nice work !! How can I connect with you to get a project done.
Hey there. Thank you. My contact email is on my channel details.
Amazing work! Could you integrate text to speech and speech to text?
Sure! Thats possible as well.
@@KieCodes Could you create a video about that?
Good idea. I put it on my list. Thank you.
Will you show how you integrated with your website?
Hey. What do you mean? The code is available for you on GitHub.
Do you know how i can connect the assistant with my database?
Yes you need to add functions for the different database queries you want to do. Like get_user_account_data(username) and this will then handeld by you backend software and be handed over to the assistant to give a proper answer to the user.
Help me pls
Sure. What’s the problem?
Very fucking cool! Tku🦾🦾🦾
Thank you very much!!!