News
Chinese AI Startup Shengshu Launches Vidu Q2, Challenging OpenAI's Sora and Google's Veo
Chinese AI Startup Shengshu Launches Vidu Q2, Challenging OpenAI's Sora and Google's Veo
Chinese AI startup Shengshu AI launches Vidu Q2, a new video generation model designed to compete with OpenAI's Sora and Google's Veo, featuring advanced 'Reference-to-Video' capabilities.
Chinese artificial intelligence startup Shengshu AI has unveiled Vidu Q2, a new video-generation model designed to compete with industry leaders like OpenAI's Sora and Google's Veo. This development intensifies the global race to create more realistic and controllable AI-generated videos, with Shengshu aiming to offer advanced features at competitive pricing.
Key Takeaways
Shengshu AI has launched Vidu Q2, a new AI video generation model.
The model aims to challenge OpenAI's Sora and Google's Veo.
Vidu Q2 features enhanced "Reference-to-Video" capabilities, allowing multiple image inputs for greater control.
The platform offers global access via its API and tiered membership plans.
Vidu Q2's Advanced Features
Shengshu's Vidu Q2 introduces a "Reference-to-Video" feature that allows creators to use up to seven reference images to define specific elements like faces, gestures, scenes, or props. This capability enables the merging of disparate elements—such as distinct characters, objects, and backgrounds—into a single, coherent video clip. The company claims this significantly improves realism, cinematic motion, and prompt comprehension.
Luo Yihang, CEO of Shengshu Technology, stated that Vidu Q2 "marks a new chapter in AI video creation," moving beyond basic video generation to "teaching AI to act and tell stories alongside creators." The model supports reference-to-video, image-to-video, and text-to-video generation, producing clips up to eight seconds long in various aspect ratios. It also includes synchronized dialogue and sound effects.
Global Accessibility and Pricing
The Vidu Q2 model and its application programming interface (API) for enterprises are available globally through Shengshu's platform. The platform offers various membership tiers, ranging from free access to a flagship plan costing up to 600 yuan (US$84) per month. The premium plan provides faster output and increased generation capacity.
Competitive Landscape
Founded in March 2023, Shengshu AI is based in Beijing and focuses on multimodal AI models through its Vidu platform. The company has previously released Vidu 1.5, Vidu 2.0, and Vidu Q1. Shengshu's user base has reportedly exceeded 30 million across more than 200 countries, with annual recurring revenues surpassing US$20 million within eight months of launch.
The release of Vidu Q2 comes amid a fierce global competition in AI video generation. OpenAI introduced Sora 2 in late September, emphasizing its improved physical accuracy, realism, and controllability. Google DeepMind unveiled its "general-purpose world model" Genie 3 in August and recently updated its Veo 3.1 video generator. In China, other notable entrants include Kuaishou's Kling, ByteDance's Jimeng, Alibaba's Wan, and Tencent's Hunyuan series model.
Sources
Chinese AI start-up Shengshu unveils Vidu Q2 in challenge to OpenAI’s Sora, South China Morning Post.
Chinese AI start-up Shengshu unveils Vidu Q2 in challenge to OpenAI's Sora, Yahoo! Tech.
China's ShengShu Updates AI Video Generator to Rival Sora, Veo, Yicai Global.
