Quantization-Aware Healing: a compressed, 4-bit model that outperforms its full-precision original

T1Covered by 1 source · first reported by Hugging Face Blog at 25 Aug, 11:39 UTC

Sources

• T1Quantization-Aware Healing: a compressed, 4-bit model that outperforms its full-precision originalHugging Face Blog25 Aug, 11:39