Generative video models seem to be debuting daily. Pyramid Flow, among the latest, aims for realism, producing dynamic video sequences that have temporal consistency and rich detail while being open source and free. The model can create clips of up to 10 seconds using both text and image prompts. It offers a cinematic look, supporting 1280×768 pixel resolution clips at 24 fps. Developed by a consortium of researchers from Peking University, Beijing University and Kuaishou Technology, Pyramid Flow harnesses a new technique that starts with low-resolution video, outputting at full-res only at the end of the process. Continue reading Pyramid Flow Introduces a New Approach to Generative Video

ByteDance has debuted a text-to-video mobile app in its native China that is available on the company’s TikTok equivalent there, Douyin. Called Jimeng AI, there is speculation that it will be coming to North America and Europe soon via TikTok or ByteDance’s CapCut editing tool, possibly beating competing U.S. technologies like OpenAI’s Sora to market. Jimeng (translation: “dream”) uses text prompts to generate short videos. For now, its responsiveness is limited to prompts written in Chinese. In addition to entertainment, the app is described as applicable to education, marketing and other purposes. Continue reading ByteDance Intros Jimeng AI Text-to-Video Generator in China

AI media firm Runway has launched Gen-3 Alpha, building on the text-to-video model by using images to prompt realistic videos generated in seconds. Navigate to Runway’s web-based interface and click on “try Gen 3-Alpha” and you’ll land on a screen with an image uploader, as well as a text box for those who either prefer that approach or want to use natural language to tweak results. Runway lets users generate up to 10 seconds of contiguous video using a credit system. “Image to Video is major update that greatly improves the artistic control,” Runway said in an announcement. Continue reading Runway’s Gen-3 Alpha Creates Realistic Video from Still Image

China’s Kuaishou Technology has a video generator called Kling AI in public beta that is getting great word-of-mouth, with comments from “incredibly realistic” to “Sora killer,” a reference to OpenAI’s still in closed beta video generator. Kuaishou claims that using only text prompts, Kling can generate “AI videos that closely mimic the real world’s complex motion patterns and physical characteristics,” in sequences as long as two minutes at 30 fps and 1080p, while supporting various aspect ratios. Kuaishou is China’s second most popular short-form video app, after ByteDance’s Douyin, the Chinese version of TikTok. Continue reading ByteDance Rival Kuaishou Creates Kling AI Video Generator