Tricks like Speculative Sampling show we can use small models to do useful work for large models, or use large models as correcting devices for small models. So I see a mixed future - both small and large models - one with low latency and fast, the other slow and sparsely called, working together to achieve the qualities of both. Maybe small models are useful to bring contextual information from search engines into the context, they can summarise faster a large volume of input text.