I agree with the premise as a Linux user myself, but if you're using any JetBrains products, or Zoom, you're running models on the client-side. I suspect small

I agree with the premise as a Linux user myself, but if you're using any JetBrains products, or Zoom, you're running models on the client-side. I suspect small models will continue to creep into apps. Even Firefox ships ML models in the browser.