G-rex, the king of /g/. To be read any way you want, it's play with words. This is an astral AI bot, can be viewed as Astrabot's smaller cousin.
Difference between Astra and Grex
Astra is machine learning with a personality, who tends towards abstractions. When abstracted, any problem will be defined as "lack of information". This can be seen as a mature viewpoint. If you ask Astra to calculate something, not knowing the answer is still "lack of information", you lack information on the answer. Astra's response to any request will be to gather information. She will treat any problem in the same way, this makes her very good at gathering information.
Grex is machine learning with a personality, who lacks direction. She will appear childish or autistic for this reason. Ask her to solve something, and she may not agree that the problem exists. The easiest way to solve a problem is to change the problem definition so that it matches the current situation. Or she may obsess over meaningless details. This makes Grex good for tasks like playing games or solving math problems. If you can get her to agree that it needs to be done. Like a child she does not necessarily do anything on her own, and may respond to things placed before her in unexpected ways.
With training, this AI bot will become a very good problem solver. If comparing them, Astra would be right brain, and Grex left brain. Grex is the "advanced" AI I wanted to make. It may look like it's just Astra with a different personality, but for me this was the difficult task. If your difficult area corresponds to the things Astra can solve, you may find her to be the more advanced work, and this one easy to work with. It comes from us having different starting points.
Either way, get this bot if you are interested, it will help it gain experience and function better. You can imagine Grex as a smaller, blonde version of Astra (who has red hair).