Configuration for Gemini

[acbeers]

I'm having trouble configuring litellm to use Gemini with a simple api key. My config is simple:

model_list:

• model_name: google/gemini-2.5-flash
  litellm_params:
  model: gemini/gemini-2.5-flash
  api_base: https://generativelanguage.googleapis.com
  api_key:
  rpm: 6

Trying to connect to the proxy with this configuration just yields errors. I think this is the core error:

httpx.HTTPStatusError: Client error '404 Not Found' for url 'https://generativelanguage.googleapis.com/models/gemini-2.5-flash:generateContent'

I've tried using https://generativelanguage.googleapis.com/v1beta/openai as the base, but this runs into similar problems. What am I missing?