1 paper
Jin Xu, Zhifang Guo, Hangrui Hu +35
We present Qwen3-Omni, a single multimodal model that, for the first time, maintains state-of-the-art performance across text, image, audio, and video without any degradation relat…