Luma AI releases Ray2 video generator with better physics

Luma AI made waves with the launch of its generative AI video creation platform Dream Machine last summer.

Of course, while that was only seven months ago, the AI ​​video space has been growing rapidly with the release of many new AI video creation models from rival startups in the US and China, including Runway, Kling, Pika 2.0, Sora from OpenAI, Google. Veo 2, Hailuo from MiniMax, and open source alternatives such as Hotshot and Mochi 1 from Genmo, to name a few. Even Luma itself recently updated its Dream Machine platform to include new still generation and brainstorming boards, and also launched an iOS app.

But the updates continue: today, San Francisco-based Luma released Ray2, its latest video AI generation model, available now via its Dream Machine website and mobile apps for paid subscribers (for starters).

The model offers “consistent, fast and natural movement and physics,” according to the co-founder and CEO of Luma AI. Amit Jain on his X accountand was trained with 10x more compute than the original Luma AI video model, Ray1.

“This skyrockets the success rate of production-ready generations and makes video storytelling accessible to many more people,” Jain added.

Luma Dream Machine Web Platform offers a free tier with 720 pixel generations capped at a varying number each month: Paid plans start at $6.99 per month: from “Lite”, which offers 1080p visuals, to Plus ($20.99/month) , including Unlimited ($66.49/month) and Business ($1,672.92/year).

A leap forward in the video generation

Currently, Luma’s Ray2 is limited to text-to-video conversion, allowing users to enter descriptions that are transformed into 5- or 10-second video clips.

The model can generate new videos in seconds, although at the moment it may take a few minutes at a time due to high demand from new users.

Examples shared by Luma and early testers of its Creators Program showcase the model’s versatility, including a man running through an Antarctic snowstorm surrounded by explosions and a ballerina performing on an Arctic ice floe.

Impressively, all movements in the sample videos appear realistic and smooth – and often, with subjects moving much faster and more naturally than videos from competing AI generators, which often appear to generate in slow motion.

The model can even create realistic versions of surreal ideas such as a giraffe surfingas user @JeffSynthétisé demonstrated. “Ray 2 is the real deal,” he says written the.

Other AI video creators who have tried the new model largely seem to agree, with Jerrod Lew publishes on X: “Improved cinematography, lighting and realism have arrived and it’s awesome.”

“…it’s so good!” Heather Cooper, AI video artist intervened.

My own testing was mixed, with more complex prompts creating unnatural and problematic results. But when it produced clips that looked more like what I had in mind in my prompts, like fencers crossing swords aboard a space station orbiting Jupiter – it was undeniably impressive.

Jain said Luma will also add image-to-video, video-to-video and editing capabilities to Ray2 in the future, expanding the tool’s creative possibilities.

To celebrate the launch of Ray2, Luma Labs is hosting the Ray2 Awards, offering creators the chance to win up to $7,000 in prizes. These include:

1. A reward on a large scale: The creator whose Ray2 content garners the most views on a single platform during the first week of launch will win $5,000. Submissions must be submitted by January 22, 2025.
2. A $3,000 draw: Creators can participate by sharing Ray2 content on social media and interacting with the Luma AI launch video. The deadline for participation is also January 22.

The winners of both prizes will be announced on January 27. Submissions can be uploaded via forms provided by Luma Labs, and creators are encouraged to use the hashtags #Ray2 and #DreamMachine when sharing their work.

Additionally, Luma Labs launched an affiliate program, allowing participants to earn commissions by promoting its tools.