CoolFace
Modelpublic

jkim96/gemma-4-26B-A4B-it-DASHQ-INT3-g64

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face
0likes47downloads
11 commits on main
b701f652mo ago

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

jkim96
61f83872mo ago

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

jkim96
945cceb2mo ago

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

jkim96
c272d212mo ago

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

jkim96
d618cd92mo ago

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

jkim96
cc377293mo ago

Standardize model card (DASH-Q)

jkim96
309f4b44mo ago

Set base_model_relation: quantized

jkim96
49c8b834mo ago

Update evaluation results

jkim96
a8425b34mo ago

Update evaluation results

jkim96
966b4f35mo ago

Upload dashq quantized checkpoint (INT3, g64, scale_zero_dtype=float16, zero-shot avg=49.4)

jkim96
7eb92265mo ago

initial commit

jkim96