3 ms·
From my experience these large context are just techinical feasability but there needs to be a lot better internal benchamarks to claim it really works. I've te
by machiaweliczny 2y ago
From my experience these large context are just techinical feasability but there needs to be a lot better internal benchamarks to claim it really works. I've tested on real world task and it all fails so far.