@chamath These CEOs will pretend they drink the AI koolaid but will never actually take the risk to seriously rethink how their company works. They are looking at AI through the lens of headcount and cost savings rather than transformative
@xai@elonmusk@milichab@JasonBud Can you guys rollout a grok-4.3-lite? $0.2/$0.5 Mtok price point is killer for enabling various workflows that dont justify 5x higher costs. Without this price point Id have to start rationing tokens pretty aggressively.
@BrianAtlas For adult women in the United States, 5'8" (68 inches or ~173 cm) is approximately the 95th–96th percentile.
https://t.co/BGD0ImolR7 - Page 20.
@PalmerLuckey@MinuteofZombie I have never seen more people wish death upon your work than on this platform. It makes no sense to me. How is beating huge contractors a problem?
@elonmusk@mark_k@xai Huge models are great but if no one can afford the tokens it doesn’t matter. This was the best feature about grok-4-1-fast-reasoning!
@steipete Is there any hope of fixing the ferocious token usage of OpenClaw? 😩😩😩 I tried it back in early Feb and it was just too expensive to actually use. Has it gotten better now?
@mcuban Doesnt address the problem of why healthcare is so expensive. Figuring out ways to help Americans "pay" isnt the right answer. The right answer is to aggressively attack the reasons healthcare is so expensive.
I find it hard to take @Anthropic's bluster seriously when their service goes down all the time, and Opus gets nerfed every 20 seconds. Maybe if their services were bullet proof and Opus was always the best, then you would have some credibility.
@elonmusk@testerlabor@elonmusk please please please, still release highly capable models like grok-4-1-fast-reasoning at the $0.2/$0.5 Mtok price, alongside your bigger models. I just cant justify the token spend!