{"ID":6023318,"CreatedAt":"2026-07-08T01:00:23.257252134Z","UpdatedAt":"2026-07-10T00:55:22.603132029Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.05722","arxiv_id":"2607.05722","title":"Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding","abstract":"We introduce Nemotron-Labs-Diffusion, a tri-mode language model (LM) that unifies AR, diffusion, and self-speculation decoding within a single architecture. Trained with a joint AR-diffusion objective, Nemotron-Labs-Diffusion can switch modes to sustain high throughput across deployment settings and concurrency levels. Our study shows that (1) AR and diffusion objectives are complementary: diffusion improves lookahead planning, while AR provides left-to-right linguistic priors. (2) In self-speculation mode, diffusion drafts while AR verifies, outperforming multi-token prediction (MTP) methods in both acceptance rate and real-device efficiency. (3) A speed-of-light analysis further demonstrates diffusion's long-term potential, with up to 76.5% more tokens per forward pass than self-speculation under an optimal sampler. Scaling to 3B, 8B, and 14B parameters, our Nemotron-Labs-Diffusion family, including base, instruct, and vision-language models, consistently outperforms state-of-the-art open-source AR and diffusion LMs in both accuracy and speed. For example, Nemotron-Labs-Diffusion-8B decodes 6x more tokens per forward than Qwen3-8B with comparable accuracy, translating to 4x higher throughput on SPEED-Bench with SGLang on a GB200 GPU.","short_abstract":"We introduce Nemotron-Labs-Diffusion, a tri-mode language model (LM) that unifies AR, diffusion, and self-speculation decoding within a single architecture. Trained with a joint AR-diffusion objective, Nemotron-Labs-Diffusion can switch modes to sustain high throughput across deployment settings and concurrency levels....","url_abs":"https://arxiv.org/abs/2607.05722","url_pdf":"https://arxiv.org/pdf/2607.05722v1","authors":"[\"Yonggan Fu\",\"Lexington Whalen\",\"Abhinav Garg\",\"Chengyue Wu\",\"Maksim Khadkevich\",\"Nicolai Oswald\",\"Enze Xie\",\"Daniel Egert\",\"Sharath Turuvekere Sreenivas\",\"Shizhe Diao\",\"Chenhan Yu\",\"Ye Yu\",\"Weijia Chen\",\"Sajad Norouzi\",\"Jingyu Liu\",\"Shiyi Lan\",\"Ligeng Zhu\",\"Jin Wang\",\"Jindong Jiang\",\"Morteza Mardani\",\"Mehran Maghoumi\",\"Song Han\",\"Ante Jukić\",\"Nima Tajbakhsh\",\"Jan Kautz\",\"Pavlo Molchanov\"]","published":"2026-07-07T01:09:54Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"Diffusion Model\",\"Language Model\"]","has_code":false}