OpenAI and Anthropic are both fiercely taking each other on. Earlier in the week, OpenAI was taunting Anthropic and delivering reset after reset after reset. Now Anthropic have joined in with their own. I am now back at 2% of weekly use, after I blew through 100% this morning. Love the competition but it comes at cost of predictability.
A domestic AI companion does not need to walk everywhere. It just needs a plan for the stairs.
A house is mostly a collection of flat surfaces joined by one deeply inconvenient invention: the staircase.
That is the obvious problem with giving an AI companion wheels. Wheels are wonderfully efficient until the floor suddenly rises by 18 centimetres. Then they become a firm commitment to remaining downstairs.
The obvious response is to give the robot legs. Humans manage stairs with two of them and robot dogs can climb remarkably well.
But walking is an expensive way to cross a kitchen floor.
There are films that are big because of their budgets, their casts or their marketing. Christopher Nolan’s The Odyssey is big in the more literal sense.
The five-hour meter is still at 26 per cent. The weekly all-models meter is at 100.
It is Thursday morning and I have just blown through my Claude Cowork weekly quota. It resets on Saturday at 6:00 a.m., so I have two days to contemplate the limit of my enthusiasm for agentic AI.
Fortunately, this is not the hard stop it once might have been. Claude simply moves to usage credits. There is no ceremony, no downgrade and no interruption to the work. I have US$25 sitting there and, as the screenshot shows, have spent US$4.06 of it already.
Beni’s all-terrain camera robot and OpenAI’s rumoured portable AI speaker point toward the same wonderfully obvious idea: a companion that follows you.
There is a very particular kind of future gadget that, once you see it, makes everything else look strangely stationary.
Nobody wants to be a model selector.
Most people do not want to decide whether a task deserves Fable, Sol, Opus, Sonnet or Haiku. They want the job done properly, quickly and without the invoice arriving as a small technical mystery.
That is where Token Blending comes in.
The idea is straightforward. Start an agentic task with the biggest, smartest and most expensive model. Let it understand the problem, make the difficult calls and create the plan. Then hand the defined work to a cheaper, faster model to execute.
The same code file now costs 30% more to process than it did three months ago. The price per token didn’t change. The tokenizer did.
Anthropic’s post-4.6 models - Sonnet 5, Opus 4.8, Fable 5 - use a new tokenizer that cuts text into more pieces than the 4.6 generation did, and you pay per piece. A detailed analysis by Playcode, who measured every frontier tokenizer on identical files, found the new Anthropic tokenizer produces 1.36 to 1.73 times GPT’s token count on the same content. TypeScript is the worst case at 1.73x.
How is it that Aesop figured this out 2,600 years ago but this wisdom is ignored in favour of “Greed is Good II”?
A CERTAIN MAN had the good fortune to possess a Goose that laid him a Golden Egg every day. But dissatisfied with so slow an income, and thinking to seize the whole treasure at once, he killed the Goose, and cutting her open, found her - just what any other goose would be!
Moral: Much wants more, and loses all.
Aesop 620 B.C. - 560 B.C.
Source: Aesop’s Fables Copyright 1881 WM. L. Allison, New York
Can’t remember where I first spotted this image. It makes a good point.
