Publication date: 16 September 2026
Creatify Labs launched Boreal on 15 September 2026, a text-to-video and image-to-video model built on Lightricks’ open LTX-2.5 base. It is available through Creatify and fal, with synchronized audio and launch pricing from $0.01 per generated second at 720p.
For advertising teams, the decision fact is iteration cost. A five-second 720p clip costs about $0.05, making it cheap enough to generate many variants, but Creatify itself says the model is strongest on short, simple scenes.
Boreal is a post-trained ad model, not a new foundation architecture
Creatify says Boreal post-trains the 22B LTX-2.5 video model on advertising footage while retaining joint audio and video generation. It supports text prompts, reference images and optional audio through the fal endpoint.
The model targets product shots, UGC-style creative and short social clips. Fal exposes commercial use and currently lists 720p, 1080p and 2K pricing, with native synchronized audio in the output.
The price advantage is large enough to change creative testing
Fal lists Boreal at $0.01 per second at 720p, $0.03 at 1080p and $0.12 at 2K. Creatify’s launch comparison puts a five-second Boreal clip near $0.05, versus roughly $0.50 for Veo 3.1 Fast, $0.56 for Kling 3.0 Pro and $0.60 for Seedance at 720p.
| Model | Approx. 5s cost | Creatify positioning |
|---|---|---|
| Boreal | $0.05 | Text or image to video |
| Veo 3.1 Fast | $0.50 | Text to video |
| Kling 3.0 Pro | $0.56 | Text or image to video |
| Seedance 720p | $0.60 | Text or image to video |
These are Creatify’s approximate list-price comparisons, not an independent cost study. The gap is still large enough that marketers can test more variants for the same budget.
The benchmark is useful, but small and vendor-designed
Creatify reports that human reviewers preferred Boreal to the untouched LTX base in 81% of decisive comparisons across 40 cases. Its launch benchmark uses 20 advertising prompts and defines a pass as beating or tying a human-made reference.
The company also discloses where Boreal is weaker. Short clips and simple scenes are the strongest use case. Single-person talking clips show the narrowest improvement over the base model. Independent benchmark coverage is not yet mature enough to validate the broader quality claims.
What marketing teams should test first
Start with a product or UGC concept that normally needs many variants. Generate the same five-second brief across Boreal and the team’s current video model, then compare usable-output rate, product fidelity, audio quality, latency and total cost.
Boreal is worth testing when volume and iteration speed matter more than maximum cinematic quality. The main reason not to switch is evidence maturity, especially for longer scenes, complex motion and multi-person consistency.
What this means
Performance marketers and social teams should test Boreal now on short paid-social variants or product shots. The first test should measure cost per usable clip, not cost per generation.
For premium brand films or longer narrative work, watch rather than switch. The price is unusually aggressive, but the public quality evidence is still based mainly on Creatify’s own small launch study.
For more information, visit the official announcement of Boreal on the Creatify website.
