Highly experimental prune of GLM 4.6V from 128 experts to 112. Vision is intact as tested on GGUF format. I would greatly appreciate feedback.

Safetensors

Model size

95B params

Tensor type

BF16

F32

Model tree for blascotobasco/GLM-4.6V-112E

Base model

zai-org/GLM-4.6V

Finetuned

(6)

this model

Quantizations