This is an attempt to use multiple gemini API keys to get high rate limits, especially on larger models(e.g. the recent 2.5 pro, wich has ratelimits of only 2 RPM). A great deal of this repo may or may not be written by gemini 2.5 pro.
- clone this repo (havent pushed to gh yet, TODO)
- copy
.env.example
to.env
. - add your API keys in
.env
. - set your API endpoint to
localhost:5001
(orserver:5001
if using a different machine for proxy server & reciever)
python3 app.py