Lightricks, the corporate behind widespread inventive apps like Facetune and VideoLeap, introduced immediately the discharge of its strongest AI video technology mannequin thus far. The LTX Video 13-billion-parameter mannequin (LTXV-13B) generates high-quality AI video as much as 30 occasions sooner than comparable fashions whereas operating on consumer-grade {hardware} somewhat than costly enterprise GPUs.
The mannequin introduces “multiscale rendering,” a novel technical method that dramatically will increase effectivity by producing video in progressive layers of element. This allows creators to provide professional-quality AI movies on normal desktop computer systems and high-end laptops as a substitute of requiring specialised enterprise tools.
“The introduction of our 13B parameter LTX Video mannequin marks a pivotal second in AI video technology with the power to generate quick, high-quality movies on client GPUs,” mentioned Zeev Farbman, co-founder and CEO of Lightricks, in an unique interview with VentureBeat. “Our customers can now create content material with extra consistency, higher high quality, and tighter management.”
How Lightricks democratizes AI video by fixing the GPU reminiscence drawback
A serious problem for AI video technology has been the large computational necessities. Main fashions from firms like Runway, Pika, and Luma usually run within the cloud on a number of enterprise-grade GPUs with 80GB or extra of VRAM (video reminiscence), making native deployment impractical for many customers.
Farbman defined how LTXV-13B addresses this limitation: “The main dividing line between client and enterprise GPUs is the quantity of VRAM. Nvidia positions their gaming {hardware} with strict reminiscence limits — the earlier technology 3090 and 4090 GPUs maxed out at 24 gigabytes of VRAM, whereas the latest 5090 reaches 32 gigabytes. Enterprise {hardware}, by comparability, provides considerably extra.”
The brand new mannequin is designed to function successfully inside these client {hardware} constraints. “The total mannequin, with none quantization, with none approximation, it is possible for you to to run on high client GPUs — 3090, 4090, 5090, together with their laptop computer variations,” Farbman famous.
Inside ‘multiscale rendering’: The artist-inspired approach that makes AI video technology 30X sooner
The core innovation behind LTXV-13B‘s effectivity is its multiscale rendering method, which Farbman described as “the most important technical breakthrough of this launch.”
“It permits the mannequin to generate particulars steadily,” he defined. “You’re beginning on the coarse grid, getting a tough approximation of the scene, of the movement of the objects shifting, and so on. After which the scene is type of divided into tiles. And each tile is stuffed with progressively extra particulars.”
This course of mirrors how artists method advanced scenes — beginning with tough sketches earlier than including progressively finer particulars. The benefit for AI is that “your peak quantity of VRAM is restricted by a tile dimension, not the ultimate decision,” Farbman mentioned.
The mannequin additionally includes a extra compressed latent house, which requires much less reminiscence whereas sustaining high quality. “With movies, you may have a better compression ratio that enables you, when you’re within the latent house, to simply take much less VRAM,” Farbman added.
Why Lightricks is betting on open supply when AI markets are more and more closed
Whereas many main AI fashions stay behind closed APIs, Lightricks has made LTXV-13B absolutely open supply, out there on each Hugging Face and GitHub. This determination comes throughout a interval when open-source AI growth has confronted challenges from business competitors.
“A yr in the past, issues had been closed, however issues are type of opening up. We’re seeing actually loads of cool LLMs and diffusion fashions opening up,” Farbman mirrored. “I’m extra optimistic now than I used to be half a yr in the past.”
The open-source technique additionally helps speed up analysis and enchancment. “The primary rationality for open-sourcing it’s to cut back the price of your R&D,” Farbman defined. “There are a ton of individuals in academia that use the mannequin, write papers, and also you’re beginning to turn out to be this curator that understands the place the true gold is.”
How Getty and Shutterstock partnerships assist remedy AI’s copyright challenges
As authorized challenges mount in opposition to AI firms utilizing scraped coaching information, Lightricks has secured partnerships with Getty Photos and Shutterstock to entry licensed content material for mannequin coaching.
“Gathering information for coaching AI fashions remains to be a authorized grey space,” Farbman acknowledged. “We now have huge prospects in our enterprise section that care about this type of stuff, so we’d like to verify we will present clear fashions for them.”
These partnerships enable Lightricks to supply a mannequin with lowered authorized threat for business functions, probably giving it a bonus in enterprise markets involved about copyright points.
The strategic gamble: Why Lightricks provides its superior AI mannequin free to startups
In an uncommon transfer for the AI business, Lightricks is providing LTXV-13B free to license for enterprises with beneath $10 million in annual income. This method goals to construct a neighborhood of builders and firms who can show the mannequin’s worth earlier than monetization.
“The considering was that academia is off the hook. These guys can do no matter they need with the mannequin,” Farbman mentioned. “With startups and business, you need to create win-win conditions. I don’t suppose you can also make a ton of cash from a neighborhood of artists taking part in with AI stuff.”
For bigger firms that discover success with the mannequin, Lightricks plans to barter licensing agreements just like how sport engines cost profitable builders. “As soon as they hit ten million in income, we’re going to come back to speak with them about licensing,” Farbman defined.
Regardless of the advances represented by LTXV-13B, Farbman acknowledges that AI video technology nonetheless has limitations. “If we’re sincere with ourselves and have a look at the highest fashions, we’re nonetheless far-off from Hollywood motion pictures. They’re not there but,” he mentioned.
Nonetheless, he sees quick sensible functions in areas like animation, the place inventive professionals can use AI to deal with time-consuming points of manufacturing. “When you concentrate on manufacturing prices of high-end animation, the true inventive work, individuals occupied with key frames and the story, is a small % of the funds. However key framing is an enormous useful resource factor,” Farbman famous.
Trying forward, Farbman predicts the subsequent frontier can be multimodal video fashions that combine totally different media varieties in a shared latent house. “It’s going to be music, audio, video, and so on. After which issues like doing good lip sync can be simpler. All this stuff will disappear. You’re going to have this multimodal mannequin that is aware of methods to function throughout all these totally different modalities.”
LTXV-13B is on the market now as an open-source launch and is being built-in into Lightricks’ inventive apps, together with its flagship storytelling platform, LTX Studio.
