Helicone Community Page

Updated last year

Caching - Helicone

Hi guys! How you doing?
I don't understand very well how caching works in Helicone. Is there anyone who can explain it better to me?
https://docs.helicone.ai/features/advanced-usage/caching

Thanks 🙂

7 comments

JJustin

Hi @funkyoz , Caching is a 1:1 cache hit. if you send the exact same request with caching turned on, you will get the same response back

ffunkyoz

Thanks @Justin Ok it’s exactly what I want to know. It’s case sensitive?

JJonny Kalambay

Is there a way I can skip caching for some requests? It's great in most cases, but with the assistants OpenAI's Assistants API it's causing me to get retrieve outdated data

JJonny Kalambay

Oh I may have answered my own question by looking at the docs, I guess I just need to omit 'Helicone-Cache-Enabled: true'

JJonny Kalambay

Actually I'm still at little stuck. If I'm using the same client across multiple routes:

Plain Text

client = AsyncOpenAI(
    base_url="https://oai.hconeai.com/v1",
    default_headers=HELICONE_HEADERS 
)

How can I change the headers after the client has been instantiated?

JJonny Kalambay

Figured it out. extra_headers param

JJustin

Oh thank you! Would you mind updating the docs? Or pointing us to where we should update the docs?

Add a reply