1 comment

[ 4.8 ms ] story [ 16.7 ms ] thread
> 12 million token context window

> 52x faster than FlashAttention at 1MM tokens

> Less than 5% the cost of Opus

This sounds too good to be true? Or am I missing something?