support windows support q4_0 and q5_0 dequant on cpu Add CopyRight from pygguf(It was added before, but disappear after merge). Add some TODO in the code.

2025-09-12 00:04:39 +00:00 · 2024-08-07 12:19:06 +08:00 · 2024-08-07 12:19:06 +08:00 · 0a2fd52cea
commit 0a2fd52cea
parent 442e13bc97
32 changed files with 248 additions and 108 deletions
--- a/ktransformers/ktransformers_ext/cuda/setup.py
+++ b/ktransformers/ktransformers_ext/cuda/setup.py
@ -10,7 +10,7 @@ setup(name='KTransformersOps',
                'custom_gguf/dequant.cu',
                'binding.cpp',
                'gptq_marlin/gptq_marlin.cu',
-                # 'gptq_marlin_repack.cu',
+               # 'gptq_marlin_repack.cu',
      ])
      ],
      cmdclass={'build_ext': BuildExtension