I was able to run OLMO 7 billion on my integrated graphics, but I was only getting nine tokens per second output, which is far too low.

The parallel holds, but one difference stands out: the 90s web rewarded early tinkerers, while LLMs are being pushed top-down by every company at once. Curious whether "left behind" will even be an option this time, or just the default fluency we all absorb.

Why does it mean to get left behind. Are you talking about the hamster wheel or a liberation movement. People ignored the privacy nightmare the internet created. AI is a tool. Not good not bad. Use it. Know its risk. Run it locally.