NVIDIA Releases Nemotron-Labs-TwoTower: an Open-Weight Diffusion Language Model Built on a Frozen Autoregressive Nemotron-3-Nano-30B-A3B Backbone

Nvidia has released Nemotron-Labs-TwoTower, an open-weight diffusion model designed to improve the efficiency of text generation. By integrating a diffusion-based architecture with an autoregressive backbone, the model addresses common throughput limitations found in standard token-by-token decoding processes. This release allows researchers to explore alternative generation methods using Nvidia's established Nemotron infrastructure.