Reference to Video AI Models
Keep characters, products and style consistent across shots: generate video guided by reference images, video and audio with Seedance, Kling, Wan, Veo, MiniMax H3 and more.
All Reference to Video AI Models
Preview the output style, compare capabilities, then open any model for full specs, pricing, and a live Playground.
Vidu Q3 Reference to Video Mix
Vidu Q3 Mix by Shengshu - animate still images into dynamic videos with AI. Transform photos into cinematic clips with natural motion, camera movement, and optional audio generation.
Vidu Q4 Reference to Video
Vidu Q4 Preview by Shengshu builds a 3 to 16 second video at up to 4K from up to 12 reference images and 3 voice clips, keeping characters, products, and voices consistent across the clip.
Grok Imagine Video Reference to Video
Grok Imagine Video is xAI's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.
Veo 3.1 Fast Reference to Video
Veo3.1 Fast by Google - animate still images into dynamic videos with AI. Transform photos into cinematic clips with natural motion, camera movement, and optional audio generation.
Veo 3.1 Reference to Video
Veo3.1 is Google's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.
Wan 2.6 Flash Reference to Video
Alibaba Wan 2.6 Flash reference-to-video model that places characters from reference images or videos into a new scene, with optional generated audio and multi-shot control.
Wan 2.6 Reference to Video
Alibaba Wan 2.6 reference-to-video model generating video guided by reference content.
Kling 3.0 Omni Pro Reference to Video
Kling 3.0 Omni Pro Reference to Video by Kuaishou generates a 3-10 second 1080p video that keeps characters and objects from reference images consistent.
Kling O3 4K Reference to Video
Kling O3 4K Reference to Video by Kuaishou generates a 3-15 second native 4K video that keeps characters and objects from reference images consistent with optional native audio.
Kling O3 Standard Reference to Video
Kling O3 Standard Reference to Video by Kuaishou generates a 3-15 second 720p video that keeps characters and objects from reference images consistent with optional native audio.
Kling O3 Pro Reference to Video
Kling O3 Pro Reference to Video by Kuaishou generates a 3-15 second 1080p video that keeps characters and objects from reference images consistent with optional native audio.
Happy Horse Reference to Video
Alibaba's #1-ranked Happy Horse 1.0 generates stunning 1080p videos with synchronized native audio and multilingual lip-sync, transforming text prompts or images into cinematic, true-to-life motion co...
Wan 2.7 Reference to Video
Alibaba Wan 2.7 reference-to-video model generating video guided by reference content.
Happy Horse 1.1 Reference to Video
Happy Horse 1.1 by Alibaba - animate still images into dynamic videos with AI. Transform photos into cinematic clips with natural motion, camera movement, and optional audio generation.
Gemini Omni Flash Reference to Video
Gemini Omni Flash Reference to Video by Google generates a short video with synchronized audio that keeps subjects, products, or style consistent with up to ten reference images.
Grok Imagine Video 1.5 Reference to Video
Grok Imagine Video 1.5 is xAI's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.
MiniMax H3 Reference to Video
Generate native-stereo 2K video guided by image, video, and audio references with MiniMax H3.
Wan 3.0 Prime Reference to Video
Wan 3.0 Prime by Alibaba generates up to 30-second videos with native audio from reference images, videos, and audio, keeping subjects, motion, and voice consistent with the references.
Wan 3.0 Prime Video
Alibaba Wan 3.0 Prime Video generates videos up to 30 seconds with native audio from text, a first/last frame, or reference images, videos, and audio.
Wan 3.0 Reference to Video
Wan 3.0 by Alibaba generates up to 30-second videos with native audio from reference images, videos, and audio, keeping subjects, motion, and voice consistent with the references.
Wan 3.0 Video
Alibaba Wan 3.0 Video generates videos up to 30 seconds with native audio from text, a first/last frame, or reference images, videos, and audio.
Gemini Omni Flash 1.1 Reference to Video
Gemini Omni Flash 1.1 Reference to Video by Google generates a short video with synchronized audio guided by up to ten reference images and up to three short reference videos.
MiniMax H3 Max Reference to Video
MiniMax H3 Max by MiniMax generates a 5-15 second video with native audio at 480P or 768P that keeps subjects, motion, and voice consistent with reference images, videos, and audio.
Seedance 2.0 Mini Reference to Video
Seedance 2.0 Mini by Bytedance - animate still images into dynamic videos with AI. Transform photos into cinematic clips with natural motion, camera movement, and optional audio generation.
Seedance 2.0 Fast Reference to Video
ByteDance's most advanced reference-to-video model in its fast tier delivering lower latency and cost without compromising on cinematic output, native audio, multi-shot editing, and director-level ca...
Seedance 2.0 Reference to Video
ByteDance's most advanced reference-to-video model generating cinematic video guided by reference content, with native audio, multi-shot editing, and director-level camera control for professional-gr...
Seedance 2.5 Reference to Video
ByteDance's next-generation reference-to-video model, generating video from multimodal references (images, videos, audio) and locking a character, set, and palette across a full take up to 30 seconds ...
Gemini Omni Flash 1.2 Reference to Video
Gemini Omni Flash 1.2 Reference to Video by Google generates a clip with synchronized audio guided by up to ten reference images and up to three reference videos, keeping faces, products, and style co...
HeyGen Video Reference to Video
HeyGen Video Reference to Video generates a 5-15 second video guided by reference images, videos and audio.
Kling V4 Reference to Video
Kling V4 Reference to Video by Kuaishou generates a 3-30 second video that keeps characters, objects and motion from reference images and videos.
Consistent subjects across every shot
Reference-to-video models generate a new clip guided by reference images, and on some models video and audio as well. Use them to keep the same character, product or set across a series of shots.
How to choose
- Multimodal references: Seedance 2.5, Wan 3.0 and MiniMax H3 accept images, video and audio together.
- Most used: Seedance 2.0 Reference to Video and MiniMax H3 Reference to Video have the highest run counts.
- Longer clips: Kling 4.0 and Wan 3.0 go up to 30 seconds.
Starting from one frame instead? See image-to-video models. To copy movement from a clip, see motion control.
Start using Reference to Video AI Models on SandBase
One API key and one request shape for every model on this page.
- 01
Create an API key
Sign in to the SandBase Console and create a key on the API Keys page. Keep it in an environment variable and never commit it to source control.
TERMINALexport SANDBASE_API_KEY="YOUR_SANDBASE_API_KEY" - 02
Submit a run
Send a request to the Unified Run API with model bytedance/seedance/2.0/reference-to-video. Only the model field changes between models on this page.
TERMINALcurl https://api.sandbase.ai/v1/run \ -H "Authorization: Bearer $SANDBASE_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "bytedance/seedance/2.0/reference-to-video", "prompt": "A peaceful cinematic shoreline at golden hour, rolling waves and soft foam, no people"}' - 03
Fetch the result
The response includes a run id. Poll it every few seconds until the status is completed, then download the output URL.
TERMINALcurl https://api.sandbase.ai/v1/run/$RUN_ID \ -H "Authorization: Bearer $SANDBASE_API_KEY"
Questions, answered
How do I refer to my references in the prompt?
Most models let you name them, for example "Image1" or "@Element1". Check the model page for its syntax.
How many references can I pass?
It depends on the model; the model page lists the limit for images, videos and audio.
Reference to video or image to video?
Image to video uses your image as the first frame. Reference to video uses references for identity and style, and composes a new shot.
Is it priced differently?
Each endpoint has its own price, shown on the model card.
Build with Reference to Video AI Models.
One key, consistent APIs, and usage-based pricing across every model on this page.