Add the 119 trial pages: 8-bit against bf16 on this checkpoint

Drop the load-time aside and the unrun page-set note

Add development-suite evaluation results for the 8-bit file

Name jaredpalmer/kev-4b as the only base model so the quantization list can see it

Link the 139fdd94 benchmark write-up (kev#162)

Kev-4B 139fdd94: 8-bit weights, head, and development-suite card

Update README.md

Update README.md

Kev-4B merged and quantized to 8 bits for MLX

initial commit