TritonMoE: a fused MoE dispatch kernel in Triton

TritonMoE implements the full MoE forward dispatch path with portable OpenAI Triton primitives instead of relying on custom vendor-specific kernels

In this paper is presented TritonMoE, a fused MoE dispatch kernel written entirely in OpenAI Triton that performs the complete forward pass using only portable Triton primitives. https:// arxiv.org/pdf/2605.23911