samithaj
/

GLM-4.7-Flash-MTP-4bit

glm4_moe_lite_mtp

speculative-decoding

multi-token-prediction

4-bit precision

Model card Files Files and versions

GLM-4.7-Flash-MTP-4bit / generation_config.json

Commit History

Add GLM-4.7-Flash MTP drafter (split via mlx-vlm glm4_moe_lite_mtp)

2586699
verified

samithaj commited on 4 days ago