« This 15-billion-parameter video model (developed by Alibaba) took first place in the Artificial Analysis rankings for video generation, ahead of Seedance 2.0. It generates text-to-video and image-to-video clips in 1080p with lip-sync in seven languages, all within a single unified pipeline »

« Combine studying, work, and creativity in a multi-model assistant (GPT, Gemini, Claude, DeepSeek, Nano Banana). This assistant can summarize PDFs and YouTube videos, solve math problems, and generate images, videos, and talking avatars in 17 languages »

« A single AI model to create realistic images, videos up to 20 seconds long, and native audio. Video generation includes native multilingual audio: dialogue, sound effects, and ambient sound are produced directly alongside the images, without the need for separate post-processing »