ByteDance Seed Introduces SeedRealtime: a Native Audio-Visual Full-Duplex LLM That Watches, Listens and Speaks in One Model

ByteDance's Seed team launched SeedRealtime, a unified audio-visual large language model that processes continuous multimodal streams in real time, integrating perception and interaction in one architecture. It is currently deployed in the Doubao app but is not publicly available for integration.