xAI launches Grok Imagine Video 1.5: a new stage in AI video generation
xAI has officially released its new model, Grok Imagine Video 1.5, from early access. This is not just an update, but a full-fledged image-to-video generator, with a primary focus on realism, physics of motion, and rendering speed. The technology is now available via API, and a faster version, Fast, has been launched for users.
The model, identified as grok-imagine-video-1.5, is already integrated into the Imagine API. Simultaneously, xAI has deployed the Video 1.5 Fast version, which operates on the web service grok.com/imagine, as well as in mobile apps for iOS and Android.
Key Improvements: Speed and Quality
The main performance boost is in the Fast version. While the previous model took over 40 seconds to generate, the new one completes the task in approximately 25 seconds. The video length remains the same at 6 seconds, and the resolution is 720p. Thus, xAI has achieved nearly a twofold acceleration of the workflow.
Quality improvements span several areas. Soundtrack, background atmosphere, and character dialogue are now generated in a single pass, perfectly synchronized with the on-screen action. Speech has become much clearer, and object movement is more stable throughout the clip. There are fewer deformations in the frames, and weight and inertia appear more realistic.
Through the API, developers can set a starting image, a text description of the motion, resolution, and duration. This opens up broad possibilities for content customization.
Additionally, xAI has introduced useful workflow tools: projects in the sidebar, parallel generation of multiple outputs, and convenient search within the personal material library.
Community Reaction and Requests for Improvement
Initial users who tested the model for several weeks generally gave positive feedback, calling the team's work high-quality. However, there were some downsides. Attentive viewers noticed an unfortunate typo in the product name in the promotional material.
Among the main community requests is the addition of voice references. This feature is necessary for creating characters with consistent sound, as seen in competing services. Users are also asking for the full versions of Video 1.5 and Imagine Pro to be made available not only via API but also in the app and on the website, even with limitations. Separate complaints were made about abrupt changes in the interface and concerns about stricter content moderation for adult users.
Cryptalist Expert Opinion: "Grok Imagine Video 1.5 is a confident step forward in AI video generation, especially in terms of motion physics and sound synchronization. However, to fully compete with market leaders, xAI needs not only to refine the interface and moderation but also to expand functionality, including support for voice references. Generation speed is impressive, but content quality must be uncompromising."