.. _apimoe_ep_communication: flashinfer.moe_ep: Expert-Parallel Communication ================================================ .. currentmodule:: flashinfer.moe_ep A :class:`MoEEpCommunication` dispatches each token to the ranks that own its experts and combines the expert outputs back on the token's source rank. :func:`create_communication` builds one from a :class:`BootstrapConfig`, which describes the expert-parallel group, and a backend config; the backends below run over NVLink (MNNVL) symmetric memory. .. autosummary:: :toctree: ../generated :template: autosummary/class_with_members.rst BootstrapConfig MoEEpCommunication MoEEpCommParams MoEEpDispatchResult NVLinkOneSidedAlltoAll NVLinkOneSidedConfig CakeAlltoAll CakeAlltoAllConfig NVLinkTwoSidedAlltoAll NVLinkTwoSidedConfig .. autosummary:: :toctree: ../generated create_communication