It has to be a good thing to stop here. We can focus on improving what we have right now. The whole stack of models is an amazing innovation no matter what. It shouldn't hurt if we pause here for a while and try to build on this or improve this.
It will be like StableDiffusion 1.5. This model can now run on low end devices, lots of open research use this model to build something else and inspire by this.
These LLMs can be used as a foundation to keep improving and building new things.