Lennart J. Kurzweg

The llm is misusung the tool. It shout be "query" not "q". The model doestnt read the error and fix itself zhe solution is a "smarter" model unfortunately

> Is this due to hallucination? @noggynoggy Hallucination is a broad term and by **my** definition this would not be that. I would just call this the model guessing what...

Alternatively (or even preferably?) a `hyprload.toml` manifest would be nice.

@francisjervis I am working on an implementation for that. Basically, you pass in a `manager_agent` into your crew, and it will be used in `.venv/lib/python3.11/site-packages/crewai/crew.py` to create the manager. This...

> @noggynoggy , interesting approach. How do you plan to add `manager_agent` to `Crew`? Overwrite `_run_hierarchical_process` 's`llm` ? #474

That's not a crew AI Issue. The LLM you are using is just "too stupid" to use tools correctly. I've maneuvered around this by using camelCase instead of under_scores. LLMs...

This is probably possible, but you'd have to use a custom manager as in #474 and then wrap the tools the Manager uses for delegation in code that acts a...

So, in other words, there is currently no support for other Architectures? That's unfortunate. I get the same error when I try to cross-build to build it on my desktop...

https://docs.crewai.com/core-concepts/Tools/ Are you sure it is to do with your code? I have experienced Mistral 7B being "too dumb" to use tools most of the time.

Looks like this issue was mentioned but closed in #573