Vidu has launched Vidu Q1, an upgrade to its generative video platform. ShengShu Technology is based in Beijing. The browser-based model of generative video transforms two still images, a text prompt and a five-second 1080p cinematic clip. Its “First to Last Frame” system guides motion seamlessly between unrelated frames. This allows solo creators to create transitions that were previously only available from professional VFX teams. Audio is now integrated into the workflow. Vidu Q1 can generate 48 kHz music and sound effects from text, and it supports ten-second multitrack layering. It also responds to timestamped cues. The company also said that the outputs of anime-style have improved with crisper lines and a better frame consistency. Internal benchmarks place Q1 ahead OpenAI’s Sora and Runway Gen-2 as well as Luma Dream Machine for frame coherence and prompt fidelity. Rivals still rely on external tools for audio and longer render times. ShengShu Technology, a Beijing-based startup founded in March 2023 specializing multimodal large language model and creative tools for film and advertising creators, is an AI startup that specializes in large language models. [TechNode report]
