Does prompt caching works on chatgpt-4o-latest?

Prompting
,
1

Accoring to the new prompt caches documentation, it says it supports

  • gpt-4o
  • gpt-4o-mini
  • o1-preview
  • o1-mini

And I was wondering if chatgpt-4o-latest is in the field of gpt-4o.

1 Like
2

Yes, and as a matter of fact:

On Wednesday, October 2nd, the default version of GPT-4o will be updated to the latest GPT-4o model, gpt-4o-2024-08-06.

2 Likes
3

Hi @HyperBlaze :wave:

Welcome :people_hugging: to the community!

Based on the email that OpenAI sent to members;
On Wednesday, October 2nd
the default version of GPT-4o will be updated to the latest GPT-4o model:

gpt-4o-2024-08-06

polepol-default model
polepol-default model671×599 27.6 KB

2 Likes
4

I wish it reduced input token limit and overall rate limit impact.

If I send a 75k token prompt, then send again followed by 75k of new text it fails based on 128k input token limit constraint.

Not saying that is the use case, but is how I tested it. We are prompt chaining a bunch of stuff together so have to send pretty much 99% same prompt over and over.

5

Thanks for responding! However, I think my original post might be misleading.
What I was asking is this model

image
image868×562 20.2 KB

The last one on the list.
Can I consider this also works with prompt caching?