[BUG] Received None or empty response from LLM call. #4093
Description
Description
tool call is OFTEN failing when using gemini 3 pro model.
openai models are working fine.
@agent
def morning_planning_agent(self) -> Agent:
return Agent(
config=cast(Dict[str, Any], self.agents_config)["morning_planning_agent"],
allow_delegation=False,
verbose=True,
tools=self._create_tools(),
llm=LLM(
model="gemini-3-pro-preview",
temperature=0.0,
api_key=settings.GEMINI_API_KEY,
),
)
Error:
Received None or empty response from LLM call.
An unknown error occurred. Please check the details below.
Error details: Invalid response from LLM call - None or empty.
An unknown error occurred. Please check the details below.
Error details: Invalid response from LLM call - None or empty.
Steps to Reproduce
configure gemini model with LLM()
Expected behavior
it had to return content with action and action input
Screenshots/Code snippets
see above
Operating System
Windows 11
Python Version
3.12
crewAI Version
1.6.1
crewAI Tools Version
1.6.1
Virtual Environment
Venv
Evidence
see above
Possible Solution
maybe it's better to switch to normal tool calling instead of returning content with action and action input? because LLMs are trained for normall tool calling, isn't it?
Additional context
.