An example of what this page produces.
AI Reference-to-Video Generator
Multi-reference AI video generator
Upload up to 30 reference images or 10 short video clips to guide characters, products, and visual style in a new AI video. Review each result because generated details can vary.
Use up to 30 reference images to guide a character, product, or look; generated details may still vary.
Model
New in 2.5: generate up to 30 seconds, guide motion with as many as 30 images, or start from audio alone.
Add at least one reference below. Images, videos, and audio can be used alone or combined.
Reference images (optional) 0/30
Reference videos (optional) 0/10
MP4 or MOV, 2-30s each, 30s combined, up to 200MB. Requires R2 with a public domain - the model fetches the file over the network.
Reference audio (optional) 0/10
Seedance 2.5 accepts MP3 or WAV clips, 2-30s each and 30s combined. Audio-only reference is supported.
Free credits are for a quick preview, not the final video. Use a short duration and lower resolution to check the subject, motion, and prompt. A fuller, more compelling final shot often needs a longer duration and higher supported resolution—use paid credits when your scene needs those settings.
Audio
With audio
Synchronized sound, voice and music
Your generated video appears here. Describe a shot and hit Generate.
Upload one or more reference images to guide appearance, plus reference video clips for motion and camera guidance. They do not have to become the opening frame, and generated details may still vary.
Use references when a prompt alone does not describe the subject or look well enough. A character sheet, product image, or palette board can give the provider more context, but it is not a post-production lock or a guarantee of consistency. Keep the set focused, describe the shot separately, and review the generated take before using it in a sequence.
Different types of reference material guide different aspects of the generated output.
Supply two or three clear images of the same face from compatible angles and lighting. This can guide a more recognizable character within and across clips, but identity, hair, clothing, and small details can still drift.
A product image gives the model real context for colourway, proportions, materials, and packaging. Review precise logos, labels, text, and small features because reference guidance is not pixel-level preservation.
A focused set of images can guide palette, contrast, era, lens character, and overall visual direction. Reusing that set can improve coherence, but it does not guarantee matching clips.
Image-to-video uses a required first-frame image and generates motion from that starting composition. Reference-to-video treats one or more images as visual guidance for a newly generated shot, so a reference does not have to appear as the opening frame. Use this mode when identity, product details, a location, or art direction matters more than preserving an exact starting frame.
Reference video clips add motion, pacing, camera behaviour, or the rhythm of an action; audio files can guide sound without requiring a visual reference. Seedance 2.5 accepts up to 30 images, 10 videos, and 10 audio files, but more inputs are not automatically better. Keep the set focused and compatible, because conflicting angles, lighting, movement, or sound can increase drift.
Use image-to-video when the uploaded image must be the first frame. Use reference-to-video when one or more assets should guide a new composition.
Multiple images can describe different views, objects, or style cues, while reference clips can contribute motion. All remain guidance rather than exact locks.
Model specifications
Published model limits, not marketing rounding. Every number below is enforced by the generator before your credits are reserved.
The model is ByteDance's. What FrameAI adds is what makes it usable at work: honest limits, honest billing, and output you are allowed to ship.
Text, up to 30 reference images, 10 reference videos, and 10 audio references can guide one Seedance 2.5 generation instead of being stitched together afterwards.
When audio is enabled, Seedance 2.5 can generate footsteps, room tone, impacts, dialogue, and score with the video. Review timing and content before publishing, or switch audio off when the clip needs separate sound design.
One prompt covers a 9:16 vertical cut for Reels, a 16:9 master for YouTube, and a 21:9 anamorphic frame for a title sequence. Resolution runs from 480p to 4K, so the same take holds up on a cinema screen and on a phone.
The maximum credit reservation is calculated from the selected model, duration, resolution, ratio, and video-reference input, then shown before submission. A successful video job can settle lower from provider-reported usage; confirmed failed jobs return the reservation.
Use Mini for cost-efficient prompt exploration, Fast when you want a balance of generation speed and cost, and Seedance 2.5 when its longer duration or richer reference inputs fit the shot. Switching models starts a new generation, so results may vary.
Content generated with a paid subscription or purchased credit pack may be used commercially under the Terms of Service. Content generated solely with free sign-up or promotional credits is for evaluation and personal use only.
Choose a focused reference set, describe the new shot, and review what the model carries into the result.
Upload up to 30 images, 10 clips, and 10 audio files. Keep each source focused on the character, product, scene, style, movement, or sound you want the model to consider, and remove references that conflict.
Let the references describe appearance, then use the prompt for the action, setting, and camera behaviour you want in the new shot. Name which reference should guide each important element when the set contains several assets.
Generate the shot, then check faces, products, text, lighting, motion, and style against the source material. Reuse a focused set for related clips, but review every generation independently.
This multi-reference video generator can combine character, product, scene, style, and motion guidance in one request.
Use two or three compatible views of the same character to give the model more identity context than a text prompt alone. Reuse the same focused set across campaign shots, then review each result for facial, wardrobe, and detail drift.
Combine a clean product image with a detail view or a style reference. This can guide shape, colour, materials, and art direction, while small text and exact logos still need close review.
Add a short reference clip when the generated shot should follow a camera move, pace, or action rhythm. The clip guides motion while your images guide appearance.
It is video generation that accepts source images, videos, or supported audio files as guidance. With Seedance 2.5, you can attach up to 30 images, 10 videos, and 10 audio files, then describe the shot you want to generate.
Up to 30 images, 10 video clips, and 10 audio files in one Seedance 2.5 request. Every model has its own per-request limit, so switching models changes how much you can attach. Reference media must also stay within the provider's per-file and combined duration limits. More is not automatically better - references that disagree with each other can degrade the result.
There is no fixed consistency guarantee. Two or three clear references of the same subject under compatible lighting can provide better guidance, while extreme angles, occlusion, or conflicting references can increase drift.
Image references guide appearance - identity, colour, materials, and art direction. Video references can contribute motion - camera behaviour, pacing, or the rhythm of an action. You can use both in the same request, but generated details can still vary.
Image-to-video uses one uploaded image as the required opening frame. Reference-to-video uses one or more images or clips as guidance for a new shot, so a reference does not have to appear as the first frame.
Only with that person’s consent. The model has no way to verify permission, so the responsibility sits with you. Generating an identifiable person without consent does not meet this service’s conditions for commercial use.
Each tool uses Seedance 2.5 for a different job. Move between them freely — your credits, history, and exports are shared.
Choose paid access when you need commercial-use eligibility, predictable generation costs, and credits restored after a failed or rejected job. One-time packs are available with no subscription.
