GSQ-RCO-GGUF Collection Non-uniform GGUF quantizations via GSQ + RCO: per-tensor mixed precision in standard GGUF form โข 4 items โข Updated about 11 hours ago โข 80
GSQ Collection GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling, https://huggingface.co/papers/2604.18556 โข 8 items โข Updated 4 days ago โข 16