Search papers, labs, and topics across Lattice.
MOSS-VL is an innovative open vision-language model that integrates real-time interaction capabilities, allowing it to perceive and generate language simultaneously through a gated cross-attention mechanism. The model is trained on a specialized interaction corpus that optimizes its ability to determine when to speak, remain silent, or revise its outputs, resulting in superior performance on streaming benchmarks. Notably, MOSS-VL-Realtime outperforms existing open-source models in proactive behavior tasks, achieving a significant lead in average scores across four benchmarks, particularly excelling in proactive alerting scenarios.
MOSS-VL achieves a staggering 66.0 average score in proactive alerting, far surpassing the best baseline by nearly 30 points.
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.
