Crew run fails due to "Received None or empty response from LLM call." when using Anthropic models

I’m working on building a coding team that includes both an architect and a coder. However, when the flow chain becomes longer, we encounter an error: “Received None or empty response from LLM call.” This issue happens more frequently when deployed on an AWS instance, while it occurs less often in a local environment.

1 Like

Hey , were you able to resolve this error , I am facing this as well , tried everything unable to get it solved. Lmk if you found a solution

I think I set
use_system_prompt = False
And it worked

1 Like

Not working bro , same error persists.

Edit : Made it work. LLM’s through groq offer 6000 tokens only , therefore we need to ask the agent to keep the answer precise and short. Sometimes it might work sometimes it might exceed the limit, that’s when this error arrises.