Reference and upload guidelines
What you can add to a render, how many, what size, and why some files are refused by content review.
What can I add as a reference?
Alongside your script you can hand a render some material to work from: still pictures, and on some styles short clips and audio too. How many you can add, and how large each file may be, depends on the style you pick. Here is where every style stands today.
Add references in the tool itself, before you submit. If a file sits outside these numbers, swap it or export it again to fit before you send the render.
| Style | How many | Files | Size | Length |
|---|---|---|---|---|
| Cinematic · Light | Up to 9 images, 3 clips and 3 audio clips | JPEG, PNG or WebP up to 10 MB; MP4 or MOV up to 100 MB; WAV or MP3 up to 15 MB | 300 to 6,000 px on each side, between 2:5 and 5:2 | Clips 2 to 15 seconds each, 15 seconds in total; audio 2 to 15 seconds each, 15 seconds in total |
| Cinematic · Ultra | Up to 9 images, 3 clips and 3 audio clips | JPEG, PNG or WebP up to 10 MB; MP4 or MOV up to 100 MB; WAV or MP3 up to 15 MB | 300 to 6,000 px on each side, between 2:5 and 5:2 | Clips 2 to 15 seconds each, 15 seconds in total; audio 2 to 15 seconds each, 15 seconds in total |
| Cinematic · Mini | Up to 9 images, 3 clips and 3 audio clips | JPEG, PNG or WebP up to 10 MB; MP4 or MOV up to 100 MB; WAV or MP3 up to 15 MB | 300 to 6,000 px on each side, between 2:5 and 5:2 | Clips 2 to 15 seconds each, 15 seconds in total; audio 2 to 15 seconds each, 15 seconds in total |
| Cinematic · Ultra Extended | Up to 30 images, 10 clips and 10 audio clips | JPEG, PNG or WebP up to 10 MB; MP4 or MOV up to 200 MB; WAV or MP3 up to 15 MB | 300 to 6,000 px on each side, between 2:5 and 5:2 | Clips 2 to 30 seconds each, 30 seconds in total; audio 2 to 30 seconds each, 30 seconds in total |
| Standard · Draft | Up to 1 image | JPEG, PNG or WebP up to 10 MB | Between 1:5 and 5:1 | Not applicable |
| Standard · Standard | Up to 1 image | JPEG, PNG or WebP up to 10 MB | Between 1:5 and 5:1 | Not applicable |
| Pro · Flash | Up to 6 images and 3 clips | JPEG, PNG or WebP up to 10 MB; MP4, WebM or MOV up to 100 MB | Between 1:5 and 5:1 | Clips 1 to 3 seconds each, 9 seconds in total |
Why was my reference refused?
Every reference passes a quick content review first. The review turns five things down:
- Brand logos and trademarks
- Copyrighted characters and artwork
- Real people's faces, unless they are a Digital Twin with consent
- Anything sexual, violent or hateful
- Anyone under 18
What about logos, brands, and copyrighted characters?
They are turned down even when the rights are yours. The review sees the picture and nothing else, so it cannot tell your own mark from someone else's and treats every mark and every well-known character the same way. Two things work well instead: describe the brand in words in your script or prompt, and use your own product photos shot with the mark out of frame.
Can I use a real person's face in a reference?
A real face rides through Kyndrify one way: as a Digital Twin, with that person's consent on record. A photo of a real person dropped in as a plain reference is turned down, including your own face. Make the Twin first on the new Digital Twin page, where the consent step is part of the flow, then render with it.A clip works the same way. As a reference it guides the scene, the motion and the look, and the person in it is not carried across, so a face in the clip does not come through to the render.
What file sizes and lengths are allowed?
Each style sets its own, and the table under What can I add as a reference has them side by side. The shape is the same everywhere: pictures, clips, and audio each have a format list and a file-size ceiling, pictures have a shape range and on some styles a smallest and a largest side, and anything with a running time is capped per file and across the whole set together.
- Files: JPEG, PNG or WebP up to 10 MB; MP4 or MOV up to 100 MB
- Picture size: 300 to 6,000 px on each side, between 2:5 and 5:2
- Length: Clips 2 to 3 seconds each, 9 seconds in total
Did not find what you needed? Tell us what happened.