1. 12 Jan, 2026 2 commits
  2. 09 Jan, 2026 2 commits
  3. 08 Jan, 2026 2 commits
  4. 07 Jan, 2026 1 commit
  5. 06 Jan, 2026 1 commit
  6. 05 Jan, 2026 2 commits
  7. 04 Jan, 2026 1 commit
  8. 25 Dec, 2025 1 commit
  9. 23 Dec, 2025 3 commits
  10. 22 Dec, 2025 5 commits
  11. 21 Dec, 2025 1 commit
  12. 18 Dec, 2025 2 commits
  13. 17 Dec, 2025 2 commits
  14. 16 Dec, 2025 2 commits
    • yangql's avatar
      up auto deepep · 0d3ae2fc
      yangql authored
      0d3ae2fc
    • laibao's avatar
      Marlin W16A16 MoE: 清理未用量化接口与辅助代码 · 80e8f551
      laibao authored
      移除 fuse_moe_w16a16_marlin.py 中未被调用的量化准备逻辑与相关导入
      精简 fused_experts_impl_w16a16_marlin 的未用量化参数,并同步更新 fused_moe.py 调用点
      清理 marlin_quant.py 中未被 vLLM 主路径使用的函数与 __main__ 调试入口,仅保留 W16A16 权重打包所需实现
      添加MOE_SUM_MUL_ADD、MOE_SUM优化
      80e8f551
  15. 15 Dec, 2025 4 commits
  16. 11 Dec, 2025 3 commits
  17. 10 Dec, 2025 1 commit
  18. 08 Dec, 2025 2 commits
  19. 07 Dec, 2025 1 commit
  20. 02 Dec, 2025 2 commits