"vllm/entrypoints/openai/responses/context.py" did not exist on "421125d03a110df7d49f84c7cf8ee9fa089d1dff"
  • zhuwenwen's avatar
    去除多余的w4a8参数 · 5ad884ee
    zhuwenwen authored
    增加fused moe文件中w4a8的相关修改
    fix: 修复W8A8读config路径错误,删除int8_utils.py文件
    fix: 修复W8A8INT8读config问题
    修改W4A8 以及W8A8量化量化092接口
    5ad884ee
test_block_int8.py 2.42 KB