New Open Source Qwen3-Next Models Preview Hybrid MoE Architecture Delivering Improved Accuracy and Accelerated Parallel Processing across NVIDIA Platform

· NVIDIA Corporation · Sept. 15, 2025, 1:06 p.m.
Featured NVIDIA Research Generative AI LLMs AI Models Hybrid MoE architecture Open Source NVIDIA
Summary
Alibaba released a preview of the new open-source Qwen3-Next models featuring a Hybrid MoE architecture aimed at improving accuracy and accelerating parallel processing on NVIDIA platforms.
Read full post on developer.nvidia.com →
AUTHOR
Sponsored
Zulip logo Zulip
Organized team chat for people who take work seriously. Topic-based threading keeps conversations focused.
Try Zulip
Become a sponsor →