Apriel-H1: Towards Efficient Enterprise Reasoning Models

Apriel-H1: Towards Efficient Enterprise Reasoning Models(arxiv.org)

1 points by guiriduro 152 days ago | 1 comment

guiriduro 152 days ago |

Apriel-H1-15b-Thinker-SFT uses incremental distillation from Apriel-Nemotron-15B-Thinker, selectively replacing less critical attention layers with linear Mamba blocks to reduce computational complexity while preserving reasoning quality.

guiriduro 152 days ago |

Apriel-H1-15b-Thinker-SFT uses incremental distillation from Apriel-Nemotron-15B-Thinker, selectively replacing less critical attention layers with linear Mamba blocks to reduce computational complexity while preserving reasoning quality.