ByteDance introduced five new AI models at Volcano Engine's FORCE conference, with Seedance 2.5 as the centerpiece. The video model is set to launch in early July and can generate single video clips up to 30 seconds long without any post-stitching, complete with scene changes and tempo shifts. Users can also edit generated videos after the fact while keeping the visual style and look intact. The new model can process up to 50 additional inputs at once, including reference images, audio, and more, which is useful for film scenes with multiple characters. According to the company, Seedance 2.0 now supports native 4K with 10-bit color depth. Volcano Engine also announced the Doubao 2.1 Pro language model, the Seedream 5.0 Pro image model, and the Seed-Audio 1.0 audio model. Doubao 2.1 Pro costs about 80 percent less than Claude Opus 4.6. The company said the new models are designed to offer enterprises access to advanced AI capabilities through its cloud business.
Seedance 2.5 represents an evolution in AI video generation, allowing for more complex and dynamic content creation. The model's ability to process multiple inputs simultaneously opens up new possibilities for creative workflows. Additionally, the integration of editing capabilities post-generation ensures greater flexibility for users. These advancements are part of ByteDance's broader strategy to expand its AI offerings through Volcano Engine.
ByteDance's Volcano Engine is the company's cloud business, which provides AI models as services for enterprises. The introduction of Seedance 2.5 and other models marks a significant step in the company's efforts to enhance its AI capabilities.
Source: thedecoder
