alt Hacker NewsI believe OP posted it because the new Kimi K3 has 69 KDA layers (the rest are 24 Gated MLA), I think previous large Kimi models had only MLA layers.
I believe OP posted it because the new Kimi K3 has 69 KDA layers (the rest are 24 Gated MLA), I think previous large Kimi models had only MLA layers.
It's not the same KDA as used in Kimi Linear, though.