Blog
Insights, tutorials, and news about LLMs, tokens, and AI tools
The 2026 Token Price War: A Field Guide to Not Overpaying
Five flagship-class models, million-token context windows as standard, and input prices that keep falling. The catch: pricing got weirder, not simpler. Here's how to read the new menu without getting fleeced.
Read More
No, GPT-5.5 Didn't Beat Opus 4.8 - and Mythos 5 Just Made the Chart Irrelevant
The screenshots claiming GPT-5.5 dethroned Opus 4.8 went viral on a single cherry-picked benchmark. Look at the full board and the story flips: Opus 4.8 holds the line, and Mythos 5 and Fable 5 don't just win - they leave the rest of the field a generation behind.
Read More
Washington 'Restricts' Claude Fable 5 - Three Days After Launch. Cui Bono?
Opinion: On June 12 the US government moved to restrict access to Claude Fable 5, days after its debut. The official reason is national security. A more cynical reading - and to be clear, this is speculation, not reporting - is that a 'ban' on the hottest model in tech is the best free marketing money can't buy.
Read More
The 700,000-Skill Gold Rush: How 'Agent Skills' Quietly Became the New App Store
A year ago nobody could agree on what an 'agent skill' even was. Today there are more than 700,000 of them, a leaderboard that looks like an app-store chart, and a land grab among Cursor, Claude Code and Copilot for who owns the install. Here's what actually happened.
Read More
Anthropic Just Split Its Smartest Model in Two: Meet Mythos 5 and Fable 5
On June 9, Anthropic did something no lab had done before: it shipped its most powerful model twice. Mythos 5 for the few, Fable 5 for everyone - same brain, very different leash. Here's why the split matters more than the benchmarks.
Read More