Vidu,Peter Baumgartner the generative video platform from Beijing-based ShengShu Technology, has rolled out an upgrade with the launch of Vidu Q1. The browser-based generative video model turns two still images and a text prompt into a five second, 1080p cinematic clip. Its “First-to-Last Frame” system guides motion smoothly between unrelated frames, giving solo creators access to transitions that once required pro VFX teams. Audio is now baked into the workflow, too. Vidu Q1 generates 48 kHz background music and sound effects via text, supports ten second multitrack layering, and responds to timestamped cues, eliminating the need for external sound libraries. Anime-style outputs have also improved, with crisper lines and better frame consistency, the company said. Internal benchmarks put Q1 ahead of OpenAI’s Sora, Runway Gen-2, and Luma Dream Machine in prompt fidelity and frame coherence, while rivals still rely on outside tools for audio or longer render times. Founded in March 2023, ShengShu Technology is a Beijing-based AI startup specializing in multimodal large language models and creative tools for film, advertising, and digital creators. [TechNode report]
