Gpt-4-32k => The model: `gpt-4-32k` does not exist

i am using gpt-4 API. but gpt-4-32k does not work even though it mentioned in the document.
what am i doing wrong??

here is the code:

    response = openai.ChatCompletion.create(

when i use model=“gpt-4” instead of model=“gpt-4-32k”, it works fine.


The larger context 32k token model "gpt-4-32k" isn’t currently available. You can only consume models that are available in the list from /Models endpoint


Does anyone know when gpt-4-32k will be available via the API? In our endpoint we only seem to have access to gpt-4-0613, however in the documentation it does reference that the 32k version “gpt-4-32k-0613” was released today also.

1 Like

32k has been around for awhile, but extremely limited rollout

We are processing requests for the 8K and 32K engines at different rates based on capacity, so you may receive access to them at different times.

How can I get access to 32k api? Anyone can help me? I want to input more text in one shot

I think for now it’s same process, apply to waitlist and wait. Supposedly they’ll be removing the waitlist and doing a bigger rollout in the next few months. Bottom line, no immediate way to get access.

For those who have already been granted GPT-4 API access, how to specifically request 32K engines?

1 Like

Thanks for reply. I already got access to gpt-4 api, do I have to apply again for gpt-4-32k api?

Hey all, there is not currently a waitlist for GPT-4-32k, the only way to get access is via submitting an eval: GitHub - openai/evals: Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.. When we start rolling our broader access, we will share more details.


Hey Logan, I hope all is well! I had 2 evals merged a little while ago and don’t have access to 32k. Is submitting an eval specific to the 8k GPT-4 access or for both 8k and 32k? :slight_smile:

@logankilpatrick Hey man, I paid $7 and I dont have Chat GPT 4, can I please get access.