This safetensors file is a conversion from the BF16 version.
This should offer a speed up.
In my tests, it's faster than GGUF Q8_0.
Model tree for noctrex/Wan2.2-S2V-14B-int8_convrot
Base model
Wan-AI/Wan2.2-S2V-14BThis safetensors file is a conversion from the BF16 version.
This should offer a speed up.
In my tests, it's faster than GGUF Q8_0.
Base model
Wan-AI/Wan2.2-S2V-14B