Stable AI has announced that its new foundational image-to-video model, Stable Video Diffusion (SVD), is now available on its developer platform and through its application programming interface (API), allowing third-party developers to integrate it into their own applications, websites, software, and services. "This new release provides programmatic access to state-of-the-art video models designed for various industries... We are releasing this version to provide developers with an efficient way to seamlessly integrate advanced video generation into their products," the company wrote in a blog post. For individuals and businesses seeking to build video generation capabilities into their applications, Stability's new SVD API plugin indeed offers one of the leading options in terms of quality. According to a post on Stability AI's LinkedIn page, the service includes "2-second videos generated by 25 frames and 24-frame FILM interpolation, completed in an average of 41 seconds." While this may not be sufficient for major video projects, it can certainly be helpful for creating GIFs with specific information, including memes. Other competitors in this space include Runway and Pika Labs, with the latter recently raising $55 million from Lightspeed Venture Partners and launching a new web platform for video generation and editing. However, these competitors' video generation AI models are not available through an API. You need to directly access their respective websites and applications to use them, which means that, at least for now, external developers cannot truly build applications on top of them or integrate them into their own applications. It is worth noting that Stability also plans to launch a user-facing video generator web experience, although there is currently no specific release date. The company encourages users to join the waiting list to become the first batch of users to try out this interface. First, let's take a look at what Stable Video Diffusion can do. Announced in a research preview about a month ago, Stable Video Diffusion allows users to generate MP4 videos by prompting static images, including JPG and PNG. Based on the samples shared by the company, the model performs quite well in generating the desired clips, but it is still in its early stages and can only generate short videos lasting up to two seconds. This is even shorter than the four-second clips generated by research-oriented video models. Of course, multiple video clips can be connected to form a longer video. To some extent, Stability claims that it can assist industries such as advertising, marketing, television, film, and gaming. What's more interesting is that unlike the model released last month for detection and feedback, the recently released model can generate videos in various layouts and resolutions, including 1024×576, 768×768, and 576×1024. It also includes additional features such as motion intensity control and seed-based control, allowing developers to choose between reproducible or random generation. Currently, Stability's developer platform API provides access to all company models, from the Stable Diffusion XL text-to-image generator to the new SVD model. The company also offers membership services to help customers host these models locally.