Researchers have introduced JoyAI-Video-Edit, a 16B-parameter autoregressive diffusion framework designed for real-time, open-ended video editing. Unlike traditional methods, this system enables causal generation without requiring access to future frames or a fixed video duration, ensuring long-term temporal consistency. To achieve high-performance streaming, the authors implemented three core technical innovations: chunk-wise autoregressive adaptation, Source-Anchored Distribution Matching Distillation (SA-DMD), and Long-Horizon Autoregressive Distillation. The SA-DMD component is specifically designed to preserve source fidelity during the two-step generation process, while the Long-Horizon distillation technique mitigates the accumulation of temporal drift inherent in autoregressive modeling. Empirical evaluations, including both automated metrics and human studies, indicate that the framework significantly outperforms existing streaming-based editors and maintains competitiveness with high-quality offline systems across various video lengths. In terms of computational efficiency, the system is capable of performing end-to-end 720p video editing at approximately 30 FPS when running on a single Nvidia B200 GPU, making it highly suitable for low-latency applications.