Skip to content

v0.2.8

Choose a tag to compare

@yanfeiwong yanfeiwong released this 11 Jul 07:10
· 6 commits to main since this release

Release Notes (v0.2.8)

性能与显存优化

  • 将 APOLLO+CAME 低秩更新路径融合到 CUDA 核心,大幅减少 Kernel 启动开销并优化临时张量生命周期,降低显存峰值。

Performance & Memory Optimization

  • Fused the APOLLO+CAME low-rank update path into CUDA kernels, reducing launch overhead and optimizing temporary tensor lifecycles to lower peak memory usage.