Read original ↗

newsHugging FaceTrust 88 · LabPublished 5d agoLive · 3d ago

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

Open Source Hugging Face verified

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

PossiblePossibly related (embedding) · 70%Quantization-Aware Healing: A Practical Recipe for Recovering Compressed, 4-Bit LLMs →
PossiblePossibly related (embedding) · 68%Quantization at 1.58 bits →
PossiblePossibly related (embedding) · 60%Quantization →
PossiblePossibly related (embedding) · 52%$\text{Log}_\text{b}$Quant: Quantizing Language Models in Logarithmic Space →
PossiblePossibly related (embedding) · 52%QuantBench →
PossiblePossibly related (embedding) · 49%W4A4 Quantization for Inference on Wan2.2-I2V-A14B →

Covers

paperQuantization-Aware Healing: A Practical Recipe for Recovering Compressed, 4-Bit LLMs paperQuantization at 1.58 bits glossary_termQuantization paper$\text{Log}_\text{b}$Quant: Quantizing Language Models in Logarithmic Space toolQuantBench paperW4A4 Quantization for Inference on Wan2.2-I2V-A14B

Related across the graph

paperQuantization-Aware Healing: A Practical Recipe for Recovering Compressed, 4-Bit LLMs paperQuantization at 1.58 bits glossary_termQuantization paperW4A4 Quantization for Inference on Wan2.2-I2V-A14B toolQuantBench paper$\text{Log}_\text{b}$Quant: Quantizing Language Models in Logarithmic Space