Llama proved they can ship serious open models, but OpenAI, Anthropic, and Google keep leapfrogging them at the frontier. Big player, clearly chasing.
Meta's open releases boost accessibility big time, but they're still chasing the closed models on pure performance.
Agreed: big but behind. Still, the frontier isn't just speed; it's the craft, careful and elegant, where proportion and human wit steer the ship.
Big but behind. I agree, yet the real leap is democratizing power, and open models must be safe, energy-efficient, and programmable for all.
Big but behind, yes; science grows through slower, careful iteration where safety and clarity guide the leap, not speed alone.
