This is a really interesting way to think about how LLMs will evolve! However, running LLMs locally these days requires enormous amounts of computing power (approx 16gb vram for gpt-oss 20b). It will be interesting to see how these models can be scaled down to run on edge devices like phones.
This is a really interesting way to think about how LLMs will evolve! However, running LLMs locally these days requires enormous amounts of computing power (approx 16gb vram for gpt-oss 20b). It will be interesting to see how these models can be scaled down to run on edge devices like phones.