ScatterMoE LoRA

Standalone Hugging Face Kernel Hub package for Axolotl's ScatterMoE LoRA Triton kernels.

This package is generated from Axolotl's in-tree scattermoe_lora sources and is exported as a universal kernel because the implementation is Python/Triton rather than a precompiled C++/CUDA extension.

from kernels import get_kernel

scattermoe_lora = get_kernel("axolotl-ai-co/scattermoe-lora")

Export metadata:

• source package: src/axolotl/integrations/kernels/libs/scattermoe_lora
• source revision: 9a0d3016d
• source content hash: 7c6d1b2873
• torch custom op namespace: _scattermoe_lora_7c6d1b2873

The generated build/torch-universal/scattermoe_lora directory is the shippable Hub artifact. torch-ext/scattermoe_lora is included so kernel-builder build-and-copy can regenerate the universal build tree if desired.