Amazon is leveling up its generative video game. The company announced a major upgrade to its AI video model, Nova Reel, which can now generate videos up to two minutes long, a notable leap in a space dominated by rapid innovation from rivals like OpenAI and Google.
Initially launched in December 2024, Nova Reel marked Amazon’s entry into the increasingly competitive world of text-to-video generation. The latest version, Nova Reel 1.1, brings improved capabilities, including multi-shot storytelling, stylistic consistency, and more granular creative control.
What’s New in Nova Reel 1.1?
According to AWS developer advocate Elizabeth Fuentes, Nova Reel 1.1 can:
-
Generate up to two-minute videos, composed of multiple six-second shots.
-
Maintain a consistent style across shots, making it more suitable for storytelling and coherent visual sequences.
-
Accept a prompt up to 4,000 characters long to guide the video generation process.
A standout feature of the update is the introduction of “Multishot Manual” mode. In this mode, users can combine a 1280 x 720 image and a 512-character prompt to generate up to 20 customizable video shots, offering more precision and creative direction.
A Growing Rival in the AI Video Arena
With the new update, Nova Reel joins the ranks of advanced generative video tools such as OpenAI’s Sora, Runway’s Gen-4 Turbo, and Google’s Veo. While it’s still early days for long-form AI-generated content, Amazon’s move signals its intent to become a serious contender in this fast-evolving domain.
The push toward longer, more coherent videos is key as brands, creators, and developers look to use AI tools for advertising, entertainment, education, and beyond.
Bottom Line
Amazon’s Nova Reel 1.1 shows that AI video generation is getting longer, smarter, and more controllable. With tech giants racing to build the next-gen media tools, the bar for creative AI continues to rise — and Nova Reel is no longer just catching up. It’s starting to compete.
AI Agents