github-featured / kvcache-ai--ktransformers

发布

0 发布 40 Git 标签

RSS 订阅源
  • v0.7.0.post1 41613064c7
    比较

    feat(GLM): add native GLM-5.3-flash support (#2173)

    frostbyte_neo 发布于 2026-08-26 14:45:12 +00:00

    • feat: support SwiGLU limit for block FP8 experts

    • perf(fp8): add native layerwise batch transport

    • feat(fp8): support layerwise transport for TP1/2/4/8

    • release: prepare GLM-5.3-flash support

    • docs(glm): refine GLM-5.3-flash documentation

    • docs(glm): simplify release announcement

    下载附件