help need to use the Gemini model while creating the agents and tasks from yaml file

[kiransair]

here is the agent configuration that I have defined in the yaml file

agent_name:
  role: >
    abc
  goal: >
    xyz
  backstory: >
    pqr
  verbose: true
  LLM: gemini/gemini-1.5-flash

as I am using the Gemini model as the llm, I have defined the Gemini API key like this (outside yaml in the code space)

import os
os.environ['GEMINI_API_KEY'] = 'api_key'

when I execute the kickoff

result = crew.kickoff(
  inputs=inputs
)

I am getting an error ERROR:root:LiteLLM call failed: litellm.AuthenticationError: AuthenticationError: OpenAIException - The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable
can someone help how to set the api_key properly while using Gemini as llm when defining agents from yaml? Do I need to define the API key in yaml?