mlboydaisuke
/

GLM-4.7-Flash-CoreAI

Text Generation

Mixture of Experts

Model card Files Files and versions

GLM-4.7-Flash-CoreAI / gpu-pipelined

Commit History

remove GatherMM int8hu — superseded by gather_qmm sym8 (2.6x, same quality)

e3abac7
verified

mlboydaisuke commited on 14 days ago

GLM-4.7-Flash: sym8 gather_qmm bundle (2.6x, clean int8)

a3f8528
verified

mlboydaisuke commited on 14 days ago

GLM-4.7-Flash: int8hu decode-pipelined bundle (ship)

5d6abab
verified

mlboydaisuke commited on 15 days ago