Honor is set to unveil the 600 series phones on April 23, bringing a major upgrade to its generative AI video tool. Last year's Image to Video feature, introduced with the 400 lineup, has seen users generate over 13.4 million seconds of content. This successor promises a vastly improved experience, moving complex processing directly onto the hardware.

The Honor 600 series will feature the first unified multi-modal video generation model designed to run entirely on the device itself. Users can input a mix of images, video clips, and text commands to direct the AI. The resulting clips will range from three to eight seconds, allowing users to dictate specific on-screen actions or fill in temporal gaps between supplied start and end frames.

To combat common AI artifacts, the new iteration boasts “multi-angle subject construction,” aiming for visual consistency even during complex simulated camera movements. Users maintain granular control, or they can opt for pre-made templates mirroring cinematic effects like the Hitchcock zoom. Access seems streamlined via a dedicated AI Button on the new hardware.

This focus on integrated, locally run AI editing tools suggests a push for immediate, complex content creation without relying on cloud processing. The Pro model will join the standard Honor 600 next week, emphasizing on-device capability over external servers for these generative tasks.