A Smarter Auto that Saves Everyone Money

There is a big gap between model capability and auto mode could be a big win if that mode were utilized to use the best thinking model to supervise the less capable auto mode ChatGPT or even ollama local models.

Whenever I see the auto model chat stream going in a direction where the model is about to do something stupid or against the rules I just provided I have to quickly stop the auto model and switch to a capable model - if I can read fast enough! Imagine if the chat stream were going at two times the speed, it would be impossible for me to intervene.

Ideally, there would be a supervisor model that was Strategy model ($$$), a Boundary model ($$) a worker auto model ($) that did the work. The Strategy model would take written (pinned requirements) documents and formulate next steps that are optimized to stay within requirement but plan next steps (these are currently the task list). Then the Boundary model would be set up to keep the worker auto in line and restrict it from going off the road.

The pinned requirements (doc’s) is a necessary step in this execution optimization. Without it, context is quickly lost.

2 Likes

Just use a Chinese model and the price solves the problem with equal or even better intelligence :slight_smile:

Intelligence is only part of the problem. The requirements soon slip out of the context window or are ignored unless they are cached and kept in cache. Models are like a child and only see what’s directly in front of them.

They need a priority memory which I guess is another way to solve this issue.

cache like Claude-4 is insane…

have you made it works the way you want?

I tried the Kim2 as well but it’s not God send or fix for keeping on track. The auto is just inferior in to Sonnet in every way. If I need to do anything right other than make some very directive changes auto is horrible. Perhaps if auto was Kim2 that might help but if you pick Kim2 you are charged like a premium model as far as I can tell.