| title | MOSS-VL Technical Report |
|---|---|
| authors | Pengyu Wang, Chenkun Tan, Shaojun Zhou, Qirui Zhou, Yanxin Chen, Xingyang He, Huazheng Zeng, Jijun Cheng, Chenghao Wang, Xiaomeng Qian, Pengfei Wang, Zhan Huang, Shanqing Gao, Wei Huang, Longjun Cao, Wu Ran, Jie Liu, Changtai Zhu, Hongkai Wang, Yixian Tian, Chenghao Liu, Zhen Ye, Xinghao Wang, Botian Jiang, Guoguo Feng, Zhaoye Fei, Ruixiao Li, Mingshu Chen, Yang Gao, Qinyuan Cheng, Shimin Li, Xipeng Qiu |
| date | 2026-08-15 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2608.15045 |
| affiliation | OpenMOSS |
| upvotes | 53 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | We present MOSS-VL, an open vision-language model family that treats real-time interaction -- perceiving while it speaks -- as a first-class capability. It is co-designed across the stack: the language decoder attends to vision only through gated cross-attention, so the model can naturally see incoming frames while generating; a synthesized interaction corpus supervises when to speak, when to stay silent, and when to revise; and a staged curriculum concentrates all real-time-specific training in one light final stage over a strong offline foundation. Offline, MOSS-VL-Instruct is competitive at comparable scale and leads temporal-reasoning video sets. Across four streaming benchmarks, MOSS-VL-Realtime posts the best average on three (second on the fourth) among open-source streaming models, sweeping the three subsets that squarely test proactive behavior -- 66.0 vs. 37.5 for the best baseline on OmniMMI Proactive Alerting. With 11.3B parameters but visual tokens outside the decoded sequence, MOSS-VL widens its time-to-first-token advantage over same-backbone Qwen3-VL-8B from 2.8x to 5.1x as visual context grows. We release all five checkpoints, the training curriculum, and the real-time inference code at https://github.com/OpenMOSS/MOSS-VL. |
cite
@misc{wang2026moss,
title={MOSS-VL Technical Report},
author={Pengyu Wang and Chenkun Tan and Shaojun Zhou and Qirui Zhou and Yanxin Chen and Xingyang He and Huazheng Zeng and Jijun Cheng and Chenghao Wang and Xiaomeng Qian and Pengfei Wang and Zhan Huang and Shanqing Gao and Wei Huang and Longjun Cao and Wu Ran and Jie Liu and Changtai Zhu and Hongkai Wang and Yixian Tian and Chenghao Liu and Zhen Ye and Xinghao Wang and Botian Jiang and Guoguo Feng and Zhaoye Fei and Ruixiao Li and Mingshu Chen and Yang Gao and Qinyuan Cheng and Shimin Li and Xipeng Qiu},
year={2026},
eprint={2608.15045},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2608.15045}
}