ERROR HTTPStatus Exception: Client error '429 Too Many Requests' for url 'https://api.openai.com/v1/chat/completions'

[ataur39n-sharif]

I am using this with Docker. Here I providing some error messages--

my command is -

docker run -it \ -e OPENAI_API_KEY=API_KEY \ -v "$(pwd)":/app zeroxeli/readme-ai:latest \ readmeai -o readme-ai.md -r https://github.com/ataur39n-sharif/book-catelog-backend

Error -

ERROR HTTPStatus Exception:
Client error '429 Too Many Requests' for url 'https://api.openai.com/v1/chat/completions'
For more information check: https://httpstatuses.com/429

images -

[eli64s]

Hi @ataur39n-sharif I just pulled the latest image and ran you're repo. Can you try once more with the latest and let me know if you still experience this?

Thanks!

[Cro22]

Same Here! I tried with the latest image.

[eli64s]

@Cro22 @ataur39n-sharif Are you using a free OpenAI account or payment method?

[Cro22]

@eli64s I use the paid OpenAI Account

[ataur39n-sharif]

@eli64s I am using a free account of OpenAI

[jatolentino]

I'm also getting the same 429 error on my readme using [https://readmeai.streamlit.app/]

[Aviksaikat]

why not add rate limiting ?

[eli64s]

@Aviksaikat There is a default rate limit setting in the config file

This seems like a common issue for unpaid accounts, but still happens for paid accounts occasionally. I may need to work on a more robust API implementation to solve this problem for everyone.

[jatolentino]

@eli64s should we increase or decrease the rate_limit variable?

[Aviksaikat]

it should be decreased.

[Aviksaikat]

how can we update the config file ?

[Aviksaikat]

that rate_limit field it totally useless.

[daan-ef2]

Got the same error and using a paid api-key. Is there any workaround?

[abhi245y]

I encountered the same issue and resolved it by switching the model to gpt-4-1106-preview. After forking the repository and reviewing the code, it appears that the issue stems from a limitation with the default gpt-4 model. The README also indicates that it uses gpt-4-1106-preview. I've implemented these changes in my local files and added a troubleshooting section to the README.

As a temporary fix, you can use the following command:

readmeai --output readme-ai.md --model gpt-4-1106-preview --repository https://github.com/eli64s/readme-ai.

However, this is my first time contributing to a project, and I'm not entirely sure about the proper procedures for contributing.

[eli64s]

@abhi245y that is correct, using the model engine gpt-4-1106-preview would be a temporary workaround. However, take note of the following before trying this.

Warning

During brief testing of the gpt-4-1106-preview model I've noticed higher API costs.
If trying this workaround, use the OpenAI API Dashboard to continuously track your API usage and cost.

Thank you,
Eli

[abhi245y]

Yah you are right, when I gave the script a few runs during testing, my usage shot up from 0.21$ to 1.4$ real quick. But if you're just using it once, it's no big deal.

I also noticed that you switched the model to gpt-4-1106-preview and then reverted it back to gpt-4. At first, I didn't understand why, but now it makes sense.

[alexiuscrow]

Had issues with —model gpt-4-1106-preview and —model gpt-3.5-turbo, but it works well with —model gpt-4.