Luma AI’s Dream Machine Lets You Generate AI Videos for Free

Luma AI launched its artificial intelligence (AI) text-to-video generation model, Dream Machine, globally on Wednesday. The platform can generate videos of up to five seconds long based on simple or descriptive text prompts. The AI ​​model can generate videos in a variety of styles, including cinematic, animation, realistic and more. The AI ​​company claims that Dream Machine is fully trained on videos and is able to generate “physically accurate, consistent, and eventful shots.” The platform is currently free to access and use, but it will likely have a daily generation limit.

Luma AI Introduces Dream Machine

According to the websiteThe Dream Machine AI model is built on a transformer model and is trained directly on videos. Typically, large language models (LLMs) are trained on text and images and then moved to videos, as they require deeper spatial and motion understanding. “Dream Machine is our first step towards building a universal imagination engine,” the company added.

Dream Machine joins video generation platforms like Runway AI and Pika 1.0, which are also available in the public domain and offer video generation ranging from three to five seconds. Gadgets 360 tried the platform and found that the platform’s prompt adherence is sub-par. It struggles with multiple characters or when the prompt is too complex. However, compared to the other two, it is capable of generating higher quality cinematic videos.

The AI ​​platform takes 120 seconds to generate a video, which the company claims will have 120 different frames. Dream Machine would also have to understand how people, animals and objects interact with the physical world, creating videos with accurate physics and character consistency.

However, Luma AI also revealed a number of limitations in the current mode, such as motion, text, morphing, and the well-known Janus problem, where the AI ​​model does not produce a single consistent 3D output, but multiple canonical representations of an object in different orientations.

Luma AI’s Dream Machine Makes a Video About Stormtroopers
Photo credit: Luma AI/Dream Machine

Luma AI didn’t share any technical details about its AI model, so information about parameter size, benchmarks, architecture, and training methods isn’t known. The company also didn’t share any details about how it procured the training data. Notably, Gadgets 360 was able to generate several videos with copyrighted characters in them.

To try out the platform, enthusiasts can go to the website and click on the ‘Try Now’ button. Users need to register for the platform before they can generate videos.