Some generalizations and fixes about autocrew HOT 2 CLOSED

The version I posted seems to consistently have a problem of reaching the max number of tokens configured into the llm, or LM Studio. I'm seeing it stop before 256 tokens. It should be possible to allow more tokens.

However, I'm thinking it might be better to restructure the output to not try to generate one whole csv for all the data. Rather, I'm going to try to generate one agent and one corresponding task at a time. Furthermore, I am thinking it might be better to try to use a crew to do all this.

from autocrew.

Thanks so much for your work on the OpenAI LLM.

I'm going to open a couple of separate issues to organise workflow better.

Feel free to fork this repo if you wish. It may be better to do this, so that you we can each issue pull requests for the code which is applicable regardless of the LLM used

from autocrew.