CoolFace
Modelpublic

av-codes/kortix-fast-apply-1.5B-v1.0-awq

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes7downloads
Model Card

AWQ version of the Kortix Fast Apply 1.5B

bash
--model av-codes/kortix-fast-apply-1.5B-v1.0-awq
--enforce-eager
--disable-log-requests
--quantization awq_marlin
--max-model-len 8192
--gpu-memory-utilization 0.25