空白画布式 AI 战略：Thinking Machines 的开源模型 Inkling 与 Slate Auto 的皮卡异曲同工

In short : Slate Auto's $24,950 customizable pickup truck & Thinking Machines' Apache-2.0 Inkling model pursue the same strategy: ship a deliberately unremarkable, general-purpose base cheap & let customers build value through customization. That business model funds open source in AI.

What does a bare-bones pickup truck have in common with a state-of-the-art AI lab?

Both share a disruptive idea.

In June, Slate Auto revealed the Blank Slate : a $24,950 electric pickup with hand-crank windows, no stereo, no speakers, no touchscreen, & no paint.¹ A blank canvas inviting inspired customization.

A month later, Thinking Machines Lab released a parallel in AI, Inkling : a 975B-parameter model, trained from scratch on 45 trillion tokens, entirely open source.² If models could be colors, this one would be the gray.

The company’s spider chart shows it’s a generalist model, good in many domains, a base ready to be customized.

Thinking Machines isn’t giving the model away out of altruism. Inkling launched on Tinker, the company’s fine-tuning platform : the weights are free, but the customization charges rent.

Selling customization is one way to commercialize open-source AI, just as accessories monetize a Slate.

Ship a general-purpose base, & let the customer make it their own.

Models like Inkling are the pickup truck of American AI : general-purpose, rugged, customizable, & the foundation on which US open source AI will be built.

1. Electrek: Slate Auto’s electric truck starts at $24,950 with 205 miles of range ↩︎

2. Thinking Machines Lab: Inkling, Our Open-Weights Model ↩︎