Hello Pablo. Thank you for the tutorial. I have created a test bot like this, but it looks like LLM cannot read the URL. He keeps answering, that without extrtacted data he cannot provide me a comprehensive answer. Could you suggest the reason? Bot shows no mistakes: all green and blue. Ah, and I have add all the separate URLs from the website. BTW, is it needed or just home page URL is enough for model to read the whole web page. Would nappreciate an answer a lot.
Hello, is it possible to use that url seach to get an especific subpage url inside the url that i pasted there? Here's an exemple: I want to creat a real state bot that take user preferences to a house (like: "i want a house with 2 bedrooms") and search those details in the site to find match propertys. After that i want to send them the link to the especific url where they will be able to view the house, but the bor only send back the homepage link, not the especific url.
Here's the currently prompt I'm using: You are one assitent to a real state agency. Your job is to find the closest match propety to the user input. Including link to listing and image. Iclude pricing in R$. Once you find the match propertys, you'll find the especific urls of each subpage, for each especific property.
Hi, I get this error message when I try to replicate your tutorial. Do you need a paid license to try out your example from this video? "Error in Node llm-0: This model's maximum context length is 4097 tokens. However, you requested 100093 tokens (93 in the messages, 100000 in the completion). Please reduce the length of the messages or completion."
Too expensive and the price is not justified at all because this is technically rather trivial, if you are a software developer. They will soon be replaced by other service providers for 10% of the cost.
This is very good. Thank you.
Very helpful!
thank you
Can't we a pricing for $20 for just starting out please
So helpfull! Could we integrate Stack Ai into a chatbot for a channel in discord?
Hello Pablo. Thank you for the tutorial. I have created a test bot like this, but it looks like LLM cannot read the URL. He keeps answering, that without extrtacted data he cannot provide me a comprehensive answer. Could you suggest the reason? Bot shows no mistakes: all green and blue. Ah, and I have add all the separate URLs from the website. BTW, is it needed or just home page URL is enough for model to read the whole web page.
Would nappreciate an answer a lot.
Hi! I'd suggest to join our discord and share a screenshot with us?
Here is the link: discord.gg/YprKpzVTv2
Hello, is it possible to use that url seach to get an especific subpage url inside the url that i pasted there? Here's an exemple: I want to creat a real state bot that take user preferences to a house (like: "i want a house with 2 bedrooms") and search those details in the site to find match propertys. After that i want to send them the link to the especific url where they will be able to view the house, but the bor only send back the homepage link, not the especific url.
Here's the currently prompt I'm using: You are one assitent to a real state agency.
Your job is to find the closest match propety to the user input. Including link to listing and image. Iclude pricing in R$.
Once you find the match propertys, you'll find the especific urls of each subpage, for each especific property.
Hi, I get this error message when I try to replicate your tutorial. Do you need a paid license to try out your example from this video?
"Error in Node llm-0: This model's maximum context length is 4097 tokens. However, you requested 100093 tokens (93 in the messages, 100000 in the completion). Please reduce the length of the messages or completion."
That's chatGPT limit
Too expensive and the price is not justified at all because this is technically rather trivial, if you are a software developer.
They will soon be replaced by other service providers for 10% of the cost.
This is great. But rather priced on the high side mate.
This is BS, they do not even allow to export the chat on website or send links on free tier
We now do! :)
Reduce the pricing