Video Generation
Video-generation models are the newest and least mature category GROUNDING tracks — text-to-video and video-editing models that extend image generation’s progress into a much harder temporal-consistency problem (objects and characters have to stay coherent across frames, not just look right in one). Progress here has been unusually fast, but production reliability and cost per generated second still lag image generation by a wide margin. GROUNDING tracks new video-model releases and the temporal-consistency/duration benchmarks that mark real progress in this category versus a flashy but narrow demo.
At a glance
- 1 tracked models
- Most recently updated: Muse Video (2026-07-08)
Most actively covered
FAQ
What is the Video Generation category?
Text-to-video and video-generation/editing models.
How many models are in this category?
GROUNDING currently tracks 1 models in Video Generation.
How current is this hub?
The most recently updated entry is Muse Video, last mentioned 2026-07-08; the radar refreshes hourly.