Where is the LLM configured?

[habsfanongit]

Sorry if this is a stupid question, but when I create a crew using the CLI, I don't see how the LLM is confirmed. All I see is the .even file with the model name and URL. I am assuming crewAI creates some default LLM using these values, but I am wondering how do I change or see the default values in Python code in the project.

In my new project using latest, I changed the model in the env file to MODEL=ollama/deepseek-r1:70b, but I want one of my agents to use the smaller version so I modified most of my agents use the lm_model property

	def researcher(self) -> Agent:
		return Agent(
			config=self.agents_config['researcher'],
			verbose=True,
            llm="ollama/deepseek-r1:7b",
		)

Is there any way, crewAI can tell me which model it used for a specific task as a debug value or something like that?

[habsfanongit]

In case anyone has similar question, I ended up using

from crewai.llm import LLM
llm = LLM(model='ollama/gemma2:27b', base_url="http://localhost:11434" )

Then using in the agent

@agent
def researcher(self) -> Agent:
	return Agent(
		config=self.agents_config['researcher'],
		verbose=True,
               llm=llm
	   )