AI 终于能边看边说了
现在的 AI 视频对话是「先看完再开口」，你说话时它得等你停，它回答时也看不见你新递过来的画面。MOSS-VL 把「边看边说」当成头等大事：它生成文字时，眼睛(视觉输入)通过一道独立的小门随时进来，不用等嘴巴(文字生成)停下来；再用合成的对话数据教它什么时候该说、什么时候该闭嘴、什么时候该改口。结果在 4 个实时对话评测里，它拿了 3 个第一，主动提醒类任务上 66.0 分，碾压最强基线 37.5 分。它不是你明天就能装进手机的东西，但这是 AI 从「问答机」走向「能陪你聊天的活人」的关键一步。
📄 原文摘要(英文)
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.