CoolFace
Modelpublic

jkim96/gemma-4-31B-it-DASHQ-INT3-g64

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face
1likes42downloads
11 commits on main
cbedc502mo ago

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

jkim96
f3c3ee62mo ago

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

jkim96
c3d40662mo ago

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

jkim96
1d692212mo ago

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

jkim96
e3aa1922mo ago

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

jkim96
d38e9643mo ago

Standardize model card (DASH-Q)

jkim96
04efb7b4mo ago

Set base_model_relation: quantized

jkim96
9955da74mo ago

Update evaluation results

jkim96
08630864mo ago

Update evaluation results

jkim96
b26021a5mo ago

Upload dashq quantized checkpoint (INT3, g64, scale_zero_dtype=float16, zero-shot avg=60.4)

jkim96
83f61c95mo ago

initial commit

jkim96