GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling Paper • 2604.18556 • Published Apr 20 • 25
November 2026 Collection Collection of Collections and Models — November 2026 • 29 items • Updated about 7 hours ago • 1
NVIDIA Blackwell-Optimized Collection Collection of NVIDIA Blackwell-Optimized Calibration and Quantizations (NVFP4, MXFP4, and More) • 29 items • Updated about 8 hours ago • 1
prithivMLmods/Ripple-65B-Preview-NVFP4-GB202-GGUF Text Generation • 65B • Updated about 10 hours ago • 8 • 1
prithivMLmods/Ripple-65B-Preview-NVFP4-GB202-GGUF Text Generation • 65B • Updated about 10 hours ago • 8 • 1
prithivMLmods/Ripple-65B-Preview-NVFP4-GB202-GGUF Text Generation • 65B • Updated about 10 hours ago • 8 • 1
prithivMLmods/Qwen3.5-4B-GLB-3D-Coder-MTP-Repacked Image-Text-to-Text • 5B • Updated about 12 hours ago • 977 • 1
prithivMLmods/Qwen3.5-4B-GLB-3D-Coder-GGUF Image-Text-to-Text • 4B • Updated about 12 hours ago • 810 • 2
MTP-Repacked Collection Collection of MTP-repacked weights for faster speculative decoding and accelerated local inference. • 6 items • Updated 1 day ago • 1