2) Open models have become tremendous value. While models like Fable/Opus 4.8 are still more intelligent, most tasks don't actually need that marginal improvement in intelligence for a much higher costs
Pretty interesting that so many different companies have basically converged to similar problems ~ the same time:
1) Frontier models are expensive. What parts of our stack can be offloaded to a dumber/faster/cheaper model (canonical example for this is subagents)