pub fn build_moe(
gpu: &dyn GpuBackend,
cfg: &Glm5NextMlpConfig,
tp_rank: usize,
full_shared_inter: usize,
load: LoadFn<'_>,
expert: ExpertFn<'_>,
) -> Result<Glm5NextMoeWeights>Expand description
2026-09-25: Bind one routed MoE site for this rank: the router and bias unsliced, the shared
expert TP-sliced, and the local_experts routed experts this EP rank owns.