Sorry that you are getting downvoted you are exactly right. People can’t handle being wrong.

This garbage analysis got deleted for a reason.

There is also zero evidence for this

Cherry picking individuals with low engagement is not a serious analysis.

To be clear four days later there is still zero evidence this is true.

There is no silence unfortunately, just a lot of wishful thinking

Apparently an unpopular opinion here but I highly appreciate the move by Apple to embrace local AI. Sure it’s not perfect in any sense, but I like the direction.

Alternative Social Media like mastodon or Bluesky are also pretty fractured

deleted by creator

No. They actually make that much money.

You can already today share a username instead of a phone number, this is just extra for signup.

This will have zero impact on the power of tech companies. Most people will just choose the algorithm, it’s just better. The only way to actually have impact is to allow people to know how their feed works and to change the algorithm.

The Chinese open source models are hosted everywhere for very cheap and they are very close to the performance of the newest models. No idea where growth for OpenAI and so on is supposed to come from

You know nothing about the situation yet you judge.

Yes it means you need help, not a murderbot

None of us is dealing with a „full set of cards“ we all depend on others to be there for us in times of need and not direct us to kill ourselves. Did you read the article?

If I buy a rig I might as well host it on the internet to get back some of the investment and sell the compute to others … wait a minute that’s cloud!

It’s always cheaper to have the same hardware serve multiple people than just one.

I also hope that don’t get me wrong, but as I said: Waiting for the LLM agent to finish coding is currently a bottleneck in software development, they don’t pay high salaries for watching the AI code, they will prefer faster agents even if they are expensive, because they are not only paying the AI Company but also the software engineer overseeing them.

I ran Gemma 4 31 B quantized so it fits in my RAM. The decoding speed was decent, but if you look at the newest models for example Gemini flash 3.5 they have a decoding speed of 280 token per second, they generate an entire page before my Mac locally generates a sentence.

Yes ofc I ran Gemma 4 for example, but compare that to the speed of Gemini in the cloud the difference is massive.