4 comments

[ 5.0 ms ] story [ 15.4 ms ] thread
They say it's infinite but then they use it to increase Llama's context length to 32k tokens instead of infinity. Still seems interesting.
Actually cool if it works. This would make many things possible on a single desktop. I'm afraid landmarking is actually a lossy compression. And the real content's size still limits model's even theoretical abilities.
It's infinite but not free. Larger context still means more VRAM used and longer compute times.