search

HKUDS / LightRAG

"LightRAG: Simple and Fast Retrieval-Augmented Generation"
https://arxiv.org/abs/2410.05779
MIT License
9.63k stars 1.19k forks source link

How to use Gemini with OpenAI Compatible #175

Closed Jaykumaran closed 3 days ago

Jaykumaran commented 3 weeks ago

Hi Team,

Can i get reference for using Gemini models with OpenAI proxy server for running my experiments with LightRAG.

Also any plans to include Gemini model example scripts anyitme soon?

Looking Forward to hear from you.

LarFii commented 3 weeks ago

We are not very familiar with the Gemini model. But examples directory offers various model methods that might assist you.

Jaykumaran commented 3 weeks ago

Hey Team,

Thanks for the response. It would be really helpful if you could explore that aspect as well. Gemini offers 1,500 free API calls per day, so it would be great to experiment with robust models like Gemini to assess performance rather than just using 2B Ollama models.

Ref: Gemini API Pricing

vanhdz2611 commented 3 days ago

@Jaykumaran Hi, are you successfully using the Gemini API ?

Jaykumaran commented 3 days ago

Hello,

Recently Gemini natively integrated OpenAI Proxy and here is the ref:

https://developers.googleblog.com/en/gemini-is-now-accessible-from-the-openai-library/