v1.15.1: Patch release
-
Relax dependency on accelerate and datasets in OVQuantizer by @eaidova in #547
-
Disable compilation before applying 4-bit weight compression by @AlexKoff88 in #569
-
Update Transformers dependency requirements by @echarlaix in #571