redlock

I hope this isn't Opus 4.5

Not true anymore since Gemini 2.5 pro

I have quizzed it with three books (total more than 1500 pages) and it gave great answers.

Initially yes when they released 2 million context with Gemini 1.5 it wasn’t effective.

Try it with Gemini 3 pro/flash now.

“But clearly the difference between LLMs in 2025 and 2023 is not as large as between 2023 and 2021.”

This is a ridiculous statement. A simple example of the huge difference is context size.

GPT-4 was, what, 8K? Now we’re in the millions with good retention. And this is just context size, let alone reasoning, multimodality, etc.

Easier to debug deterministic inference

Really? Mansplain? Why bring gender wars terms into this.

Doesn't OpenRouter ranking include pricing?

Not really a good measure of quality or performance but of cost effectiveness

I wouldn't call the inflation that ravaged the world and real estate prices beyond the means of the middle class "saving".

His quantitative easing and zero interest rate policy only saved asset managers and corporate balance sheets.

Noam Shazeer is not any guy. And I would bet the latest jump in Gemini capability is a result of him coming back.

This is my observation as well. Recent market gains doesn't reflect how inflation eroded most people purchasing power.

People today are worse off than before Covid even though the market is much higher.

Debt fueled growth (through quantitative easing and deficit spending) is not healthy and always has a bad ending.

Real economic growth is what was seen between 1950 and 1970 where purchasing power increased and most of the gains went to the middle class.

Moore's law is exponential

The issue here is that, even with a lot of VRAM, you may be able to run the model, but with a large context, it will still be too slow. (For example, running LLaMA 70B with a 30k+ context prompt takes minutes to process.)