Wan2.2-T2V-A14B INT8 ConvRot
Converted from FP16 source weights (Wan-AI/Wan2.2-T2V-A14B, via Comfy-Org repack) with silveroxides/convert_to_quant:
ctq -i <source>.safetensors -o <output>.safetensors \
--int8 --scaling_mode row --simple \
--convrot --convrot-group-size 256 \
--wan --comfy_quant --save-quant-metadata
License
Apache 2.0, inherited from the base weights (Wan-AI).
- Downloads last month
- -
Model tree for FartingBackwards/wan2.2-t2v-14B-int8-convrot
Base model
Wan-AI/Wan2.2-T2V-A14B