CoolFace
Modelpublic

jkim96/gemma-4-26B-A4B-it-DASHQ-INT3-g32

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face
0likes55downloads
11 commits on main
fcabf422mo ago

Add DASH-Q remote-code inference (Triton decode kernel)

jkim96
9d364e72mo ago

Add DASH-Q remote-code inference (Triton decode kernel)

jkim96
32ce3c32mo ago

Add DASH-Q remote-code inference (Triton decode kernel)

jkim96
3efa18a2mo ago

Add DASH-Q remote-code inference (Triton decode kernel)

jkim96
a3c231f2mo ago

Add DASH-Q remote-code inference (Triton decode kernel)

jkim96
63759ef3mo ago

Standardize model card (DASH-Q)

jkim96
e1c674a4mo ago

Set base_model_relation: quantized

jkim96
bbee53b4mo ago

Update evaluation results

jkim96
1262e2a4mo ago

Update evaluation results

jkim96
a62f1ea5mo ago

Upload dashq quantized checkpoint (INT3, g32, scale_zero_dtype=float16, zero-shot avg=48.6)

jkim96
de1236a5mo ago

initial commit

jkim96