Pinned Loading

1. MixedGemm MixedGemm Public

   a mixed-precision gemm with quantize and reorder kernel.

   Python 14 1

2. lwy2020/MicroMix lwy2020/MicroMix Public

   MicroMix: Efficient Mixed-Precision Quantization with Microscaling Formats for Large Language Models

   Cuda 22 3

3. NPU-INT8 NPU-INT8 Public

   Perform INT8 GEMM on Ascend NPUs, with per-token quantization.

   C++ 1

4. ARCQuant ARCQuant Public

   ARCQuant: Boosting Fine-Grained Quantization with Augmented Residual Channels for LLMs

   Cuda