1/5 Why is crypto weak lately? The answer may not lie in crypto itself — but in equities. CBOE Dispersion Index hit 42 — 3rd highest ever. It means extreme capital concentration within the S&P 500. When a few hot themes absorb all the flows, BTC gets sidelined. 🧵👇

India - 1.5 Billion people Africa - 1.6 Billion people US - 340 million UK - 69 million France - 66 million Spain - 49 million Canada - 41 million Sweden - 10 million Ireland - 5 million If mass immigration continues Western nations, cultures and peoples will be gone.

Fable 5.1 hasn’t even been out for 24 hours, and my whole feed is about two things: how good it is and how fast it burns tokens. I’ve already hit the limits on my accounts. Some initial thoughts. 1/ 5.1 is the strongest first impression I’ve had from any Anthropic model. It feels more deliberate than Fable 5, sometimes almost Opus-5-like. Still, I haven't gained Fable 5 like trust for some tasks. 2/ I get why people are annoyed by the limits. For what I’m doing, they’ve been manageable. We built a new design system and asked 5.1 to migrate nearly 5,000 components. It has been running since last night, with Opus 5 as fallback. We hit limits faster than Fable 5 for sure. 3/ When the first prompt is clear and detailed, I usually don’t need a second one. The logic has been excellent so far. I haven’t done enough UI work with it to judge that side yet. 4/ Fable Orchestrator + @grok is a mad combination. Fable sets the task and checks the result. Grok does the work. In my runs, the orchestrator kept pushing Grok past the first acceptable answer until the checks passed. For the first time, I hit the limit on my Grok plan. --- For me, this puts Anthropic way ahead of OpenAI right now. At least until Astra arrives. The one thing I still love about OpenAI is its multi-agent harness. I hope Astra can retain that. I doubt Astra beats Fable 5.1 out of the gate, but you never know. If it does, OpenAI will have pulled off something huge.