[METHOD] Make Your Own PAA Site With OpenAI - Complete Tutorial + FREE Python Script

Hello,

I improved your script:

1. added a cache so that downloaded responses do not "disappear" when an error pops up (e.g. API stops responding) the script creates an additional working file.
2. I added a progress bar that also calculates the estimated time.



cheers :)
Hello thank you very much! Can you share it as a code please ( With tabs ) ?
 
Anyone else getting this error?

openai.error.RateLimitError: The server had an error while processing your request. Sorry about that!
 
Anyone else getting this error?

openai.error.RateLimitError: The server had an error while processing your request. Sorry about that!

I'm sorry but I've not seen this error before. Sometimes the server is overloaded but the error for that is different.

Some other people have complained about this yesterday on the support forum. I can't share the link here but if you Google your error, it's the second one.
 
Can you Set up the whole process for money? If you interest, let's cooperate ;)

I’m not interested myself but if you start a thread in the HAF section, I’m sure you can hire someone to help you out. It takes like 5 mins to set this up.
 
@warriorsam53 , thank you for sharing this code. I am using something similar more suited to my needs but struggling to put format the relevant sections' title in H2 tags.
Would you care to share any insight on how to do that?
I have tried several prompts to no avail. That's my missing piece.

Thank you.
 
@warriorsam53 , thank you for sharing this code. I am using something similar more suited to my needs but struggling to put format the relevant sections' title in H2 tags.
Would you care to share any insight on how to do that?
I have tried several prompts to no avail. That's my missing piece.

Thank you.

Try including an example of the output you’d like to see within the prompt.

I’m sorry I can’t be more specific since I don’t want to give away the prompts I use for my own sites.
 
Try including an example of the output you’d like to see within the prompt.

I’m sorry I can’t be more specific since I don’t want to give away the prompts I use for my own sites.
Not a problem. Thanks for getting back to me. I will keep testing.
 
Anyone creating auto-blogs with this?
 
Thanks again for this awesome example. You have made the point that the key to success is in the prompts, but, like others, I'm finding that even with the token number set to 3995, I'm not getting much more than 400 words in the answer in a reliable way. So a question to warriorsam53: is it even possible to get longer answers (ca 1000 words) without multiple steps, just by clever prompt engineering? The issue seems to be that the more info you put in the prompt, the less you'll get out because of the token limit. If that's the case, fine - the script can be modified to get multiple answers from one PAA query, but it would be helpful to understand the limits from someone who's been doing this a while.
 
Thanks again for this awesome example. You have made the point that the key to success is in the prompts, but, like others, I'm finding that even with the token number set to 3995, I'm not getting much more than 400 words in the answer in a reliable way. So a question to warriorsam53: is it even possible to get longer answers (ca 1000 words) without multiple steps, just by clever prompt engineering? The issue seems to be that the more info you put in the prompt, the less you'll get out because of the token limit. If that's the case, fine - the script can be modified to get multiple answers from one PAA query, but it would be helpful to understand the limits from someone who's been doing this a while.

Yes, it's possible to get longer content (700-1200) words using a single prompt. The token count doesn't work like you think. Although restricting the number of tokens can result in shorter content, increasing the number of tokens doesn't increase the length of the output beyond a certain point.

For example, let's take a simple prompt: write a blog introduction paragraph for the topic XYZ.

Whether you set the tokens to 250 or 3500, it won't make any difference. You'll still get one paragraph.

Similarly, the number of tokens is not what's restricting your output. It's the prompt.

It takes approximately 1000 tokens to generate 750 words. But just setting 1000 as the number of tokens does not generate 750 words.

I know that it's not intuitive but that's how it works. Even with the best prompts ever, it'll be hard to generate 2000+ words using a single prompt. Once again, this is not because you run out of tokens.
 
Technically, what is the difference between a PAA website and a normal blog? Does PAA use HTML tags in a specific order?
 
You're more likely to see questions as the main title/blog posts as well as a shit-ton more bulleted lists in PAA vs blogs (especially if it's AI generated.. these bots have a hard on for breaking shit down into a 5 point list and then ending with "In conclusion."). If it's set up correctly though, there shouldn't be a difference between a well manufactured PAA site and a blog as far as <h1> and various <meta> tags are concerned.
 
Back
Top