MOSS-VL: An open VLM family that treats real-time interaction as a first-class capability

MOSS-VL技术报告：将实时交互作为一等能力的开源视觉语言模型家族

Fudan's MOSS-VL makes real-time interaction—perceiving while speaking—a first-class capability. Gated cross-attention keeps visual tokens outside the decoded sequence, giving it a 2.8× to 5.1× time-to-first-token advantage over same-backbone Qwen3-VL-8B. MOSS-VL-Realtime tops three of four streaming benchmarks, hitting 66.0 vs. 37.5 on OmniMMI Proactive Alerting. The offline variant leads temporal-reasoning video sets at comparable scale. All five checkpoints, the training curriculum, and inference code are open.

Why it matters: Fudan open-sourced a VLM family that treats real-time interaction as a first-class capability. Gated cross-attention cuts time-to-first-token by 2.8–5.1× vs. Qwen3-VL-8B on the same base, with the gap widening as frames increase. H and K are solid hits, but R is weak—the open-...