The problem I've found with all of them is it is difficult to get consistent solid results that you would want to build around. Even listening to podcasts this seems to be a theme with the people at the forefront of MAS LLMs.

Thanks to the cognitive architecture design, goal-oriented behavior is at the core of our LLM agents. When presented with a goal, they employ a planning and execution process to achieve it. If the goal can be accomplished using atomic game actions, we generate a plan that outlines how to solve the objective. For longer-term goals, we decompose the... See more

Yes, LLMs can reason , plan , make judgments under uncertainty , and integrate these skills in a wide variety of domains; but true AGI supposedly rests on the assumption that these systems, given their ability to match or surpass human intelligence , would be capable of going out and actively doing things we’re doing.