yyj
|
41613064c7
|
feat(GLM): add native GLM-5.3-flash support (#2173)
* 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
|
2026-08-26 22:45:12 +08:00 |
|