Creatify models, access and benchmark guide
Creatify model and access guide with a practical US-English evaluation brief and links to publisher materials.

👁 111 reads
A retail video brief can require a product photo, presenter and final customer-facing crop. Before quoting a Creatify workflow, check which part its API exposes and whether the chosen format keeps the real packshot intact.
Boreal’s text-to-video preference row concerns the output of a generation model. Creatify’s developer documentation describes a product workflow, where the buyer may also need an avatar, a product reference and a particular delivery format. I would make one brief for the model and one for the final advertisement. A presenter’s hand can look natural in the original frame but cover the product after the vertical crop. Check the actual deliverable before charging for a repeat generation.
Choosing the first job
Show the same cereal box throughout a vertical product demo and confirm that the side panel stays on the correct side.
Models in the November 17 snapshot
| Model and track | Placement | Reported score |
|---|---|---|
| Boreal (AA text to video) | 35 | 827 Elo |
Release record. Publisher documentation identifies the product family. The ranked configurations include Boreal.
Google · Your sourcesQuantslant higher in your Google - one clickAdd to preferred sources →Sources
- https://docs.creatify.ai/ (accessed November 17, 2026)
- https://creatify.ai/ (accessed November 17, 2026)
- https://artificialanalysis.ai/video/leaderboard/text-to-video (accessed November 17, 2026)
More on this from QUANTSLANT
ai tools·Speech synthesis ranked with the voice held constantCompare controlled-voice TTS rankings for a consistent narrator and plan a blind listening check using a consented voice.