Show HN: A 6M-token movable window on a single 46GB GPU

Discussion (6 Comments)Read Original on HackerNews

himata4113•about 1 hour ago

"No implementation detail, algorithm, or configuration is contained in this document by design."

+ odd page cuts, it's as-if no human has ever looked at this before uploading it.

SwellJoe•16 minutes ago

These LLMs are absolute poison for some folks.

stephantul•about 1 hour ago

100% generated. I skimmed the paper, and came out with a feeling of still not knowing what this is about.

Wetime•44 minutes ago

We handeld llm like a human brain we decoupled knowledge from the memory and build a memory layer that makes redoing things free and fast, so the llm can once it learned something solves it for free the next time

SwellJoe•about 2 hours ago

You couldn't be bothered to write a coherent summary of what this actually is and what it does, you just let the AI write some random noise, eh?

Wetime•about 1 hour ago

This paper show a new tool called Galahad.Normally, AI has to think and guess the answer every time, which costs time and money.

The knowledge of the model grows next to it not the model itself and no it is not the same as cache

No fine-tuning needed

It gives the exact same right answer every time, costs zero extra tokens, and saves lots of energy

SwellJoe•about 1 hour ago

So, you don't know what it is, either.

Wetime•44 minutes ago

We handeld llm like a human brain we decopelled knowledge from the memory and build a memory layer that makes redoing things free and fast, so the llm can once it learned something solves it for free the next time

Show HN: A 6M-token movable window on a single 46GB GPU

⚡ Community Insights

Discussion (6 Comments)Read Original on HackerNews