Creatify Boreal: release, price, how to use

Creatify Labs released Boreal on 15 September 2026. It makes product, UGC and presenter videos from a text prompt, or a prompt plus a photo, with speech and sound generated in the same pass. fal lists it from $0.01 per second at 720p. On Versely a clip runs from 4 to 192 credits.

Key facts

Item	Detail
Released	15 September 2026, by Creatify Labs
Does	Text and image to video, with dialogue, ambience, music and effects in one MP4
Length	fal: 1 to 20 s at 24 fps. Creatify's API: up to 60 s at 720p, 30 s at 1080p, 15 s at 2K
Resolution	720p, 1080p or 2K; auto, 16:9, 9:16, 1:1, 4:3 or 3:4
Language	English is the validated language, per fal
fal price, 16 October 2026	$0.01 per second at 720p, $0.03 at 1080p, $0.12 at 2K
Creatify API price	1, 3 or 12 Creatify credits per started 10 s at 720p, 1080p, 2K; plans from $99 a month
Where	Creatify's app and API, fal, Picsart, Versely

Creatify's launch post called the output "720p-class" and set it against Seedance 2.0 at 1080p; the 1080p and 2K tiers appear in the fal and API listings.

Sources, all read on 16 October 2026: Creatify's launch post, Labs page, API docs and billing page; fal's model page and schema; Artificial Analysis' text-to-video board. DataNorth corroborates the launch date and fal price, and Picsart the 1 to 20 second range.

What is new: Boreal versus Aurora and Boreal-H3

There is no earlier Boreal. What sets it apart from Creatify's Aurora line is the input. Creatify Labs describes Aurora 2.0, an avatar model, as turning "a single photo and one audio clip into a lifelike talking video". Boreal starts from a described scene instead: subject, action, camera, dialogue and sound. Creatify does not name the open 22B-parameter base it post-trained on advertising footage.

Creatify's API serves two models from one endpoint. Boreal is the default, listed for longer videos and for video driven by an audio track. Boreal-H3 is listed as the highest-quality option, with start and end frames and reference media, at 768p, 1088p or 2K for 5 to 15 seconds. Versely's row is the original Boreal; as of 16 October 2026 Boreal-H3 is not in the catalog. fal and Creatify's docs also document an optional audio track that drives the clip; Versely's row lists no audio input.

Checking Creatify's claims

Creatify says Boreal "rivals the quality of the top-tier closed models" at "a fraction of the price and up to 40 times the speed". Its evidence is all its own testing:

• Against its base model. Evaluators preferred Boreal in 81% of decisive blind comparisons (25 to 6, with 9 ties, across 40 production cases): 83% on product ads, 85% on creator and UGC scenes, 60% on single-person talking clips.
• Against closed models. On 20 ad prompts, a blind judge passed Boreal 70% of the time, Seedance 2.0 80%, H3 Max (MiniMax) 65% and the untouched base 50%. Creatify says Boreal "matches Seedance 2.0 on perceptual quality and leads it on motion".
• Speed. Creatify says a five-second clip takes about five seconds on its own stack, against a median 198 seconds for Seedance 2.0 in its test. Queues on hosted platforms may differ.

Independent votes are less generous. On Artificial Analysis' text-to-video board (Overall category, Elo from pairwise human preference votes) on 16 October 2026, Boreal scores 831 (95% interval 811 to 851, from 2,796 votes), joint 27th to 28th of 28 models. Wan 3.0 leads at 1,174 and Seedance 2.0 sits at 1,136. Read Boreal as a price-and-speed model, not a peak-quality one.

Scene model, avatar tool or lipsync tool

Pick by what has to stay fixed.

Tool type	You supply	You get
Scene model (Boreal)	A prompt, optionally a photo	A new scene with its own people, product, sound and speech
Avatar or photo-to-talking-video	A face and a script or audio	A presenter saying your exact words
Lipsync	Finished footage and new audio	The same footage with a new spoken line

Apart from length and resolution, Boreal's documented inputs are a prompt, an optional start image and an optional audio track: no actor library, no voice picker. You steer the person and the voice in words, as fal's own example does with "Natural confident female English voice". For a repeat face and an exact script, use an avatar tool; for new words on existing footage, lipsync. Versely's guide to the best AI models for UGC and avatar videos covers both.

Who it is for, and who should wait

Good fit:

• Teams testing many short UGC-style and product clips. Creatify says Boreal is strongest on short clips and simple scenes. Ten 5-second drafts at 720p are 40 credits on Versely.
• Product videos from one photo; Creatify pitches hero shots and demos "with the product kept intact".
• Anyone who wants speech and sound made with the picture, not synced afterward.

Test first, or wait, if:

• You need one presenter and an exact script every time. Use the UGC builder or an avatar model.
• One person talks to camera. Creatify says that is where Boreal improves least over its base, its lip sync is only level with the base, and fal says to review dialogue and lip sync before publishing.
• You need speech beyond English, the only validated language.
• Logos, small text or hands carry the shot. fal warns that these can vary.
• Peak quality matters. The independent board puts Boreal last. Creatify rates Boreal-H3 higher, and Versely does not carry it; Wan 3.0 text to video leads the board and is in the catalog.

How to use Boreal on Versely

Boreal is live on Versely as one row under Creatify: text to video and image to video, audio on, starting image optional. There is no reference, edit or video-input row. Clips run 5, 10, 15 or 20 seconds, at the resolutions and ratios above.

Length	720p	1080p	2K
5 s	4	12	48
10 s	8	24	96
15 s	12	36	144
20 s	16	48	192

Credits per clip, from a live price check on 16 October 2026: 0.8 credits a second at 720p, 2.4 at 1080p and 9.6 at 2K. The model page's grid rounds to whole credits per second; this table is the full price.

1. Open Imagine in the web app, switch to Video, choose Text to Video (or Image to Video with a product photo) and pick Boreal. The iPhone app reads the same catalog.
2. Write the prompt in Creatify's four sections, with one action, one camera move and one sound cue. Choose 9:16 for TikTok, Reels or Shorts; auto keeps a photo's own shape.
3. Draft at 720p and 5 to 10 seconds. fal's schema has no seed field, so a 1080p or 2K final is a new take, not an upscale. Ten drafts at 5 seconds are 40 credits; two 5-second 2K finals are 96 more.
4. Listen to the speech and watch the mouth before you caption.

[VISUAL]
Handheld selfie in a bright kitchen. A woman holds up a steel water bottle and taps the lid.

[SPEECH]
"Okay, this is the bottle that finally stopped leaking in my bag."

[SOUNDS]
Kitchen ambience, a light tap on metal.

[TEXT]
None.

When the presenter and script must stay fixed, the UGC ads builder works from an actor, your script and a voice, and the UGC video generator page gives the overview. Creatify's fal listing says every output carries C2PA metadata and an imperceptible watermark, but viewers cannot see either: label realistic generated people yourself, as AI UGC disclosure by platform explains.

FAQ

When was Creatify Boreal released?

Creatify Labs released it on 15 September 2026, with text to video and image to video at launch.

How much does Boreal cost?

fal lists $0.01 per second at 720p, $0.03 at 1080p and $0.12 at 2K (read 16 October 2026), so 10 seconds is $0.10, $0.30 or $1.20. Creatify's API bills 1, 3 or 12 of its own credits per started 10 seconds. On Versely a clip runs 4 to 192 Versely credits.

How long can a Boreal video be?

fal allows 1 to 20 seconds. Creatify's API allows up to 60 seconds at 720p, 30 at 1080p and 15 at 2K. Versely offers 5, 10, 15 and 20.

Is Boreal an avatar or lipsync model?

No. It generates the whole scene, speech included, from a prompt and an optional photo. Creatify's avatar line is Aurora. Boreal has no actor library, so a face that must repeat needs an avatar tool.