Seedream 5.0 Pro
High-detail text-to-image and reference-based image editing.
Seedream ProCreate professional AI images and cinematic videos. Generate with Seedream 5.0 Pro, Nano Banana Pro, and Seedance 2.0 using text, images, video, audio, and precise reference control.
Seedance 2.0 With AudioMultimodal input with powerful reference capabilitiesType @ to reference uploaded images, videos, or audio.
Generate multiple videos with the same settings, up to 10.
Choose a model for the job: high-fidelity image generation, reference-based editing, cinematic motion, or synchronized audio.
High-detail text-to-image and reference-based image editing.
Precise text rendering, multi-image composition, and controlled edits.
Multimodal video creation with image, video, and audio references.
Cinematic text-to-video and image-to-video with audio support.
Prompt-driven motion, visual realism, and synchronized sound.
Model availability and output options depend on the selected provider and generation mode.
Explore stunning video examples created with Seedance 2.0's multi-modal capabilities.














A truly controllable multi-modal AI video model. Reference anything, edit anything, create anything.
Upload up to 9 images, 3 videos, and 3 audio files. Combine text and media freely to express your vision with unprecedented flexibility.
Reference motion, effects, camera movements, characters, scenes, and sounds from any uploaded content using natural language.
Maintain perfect consistency for faces, clothing, text, scenes, and visual styles throughout your entire video.
Upload a reference video to reproduce choreography, cinematic camera movements, and complex action sequences.
Extend videos, merge clips, replace characters, and edit specific segments while preserving everything else.
Generate context-aware sound effects and background music or sync visuals perfectly to an uploaded audio track.
From viral content to professional productions, bring every multi-modal vision to life.
Create polished, high-impact video content with reference-driven motion and consistent cinematic quality.
Create polished, high-impact video content with reference-driven motion and consistent cinematic quality.
Create polished, high-impact video content with reference-driven motion and consistent cinematic quality.
Create polished, high-impact video content with reference-driven motion and consistent cinematic quality.
Create polished, high-impact video content with reference-driven motion and consistent cinematic quality.
Create polished, high-impact video content with reference-driven motion and consistent cinematic quality.
Upload images, videos, or audio as references. Combine different modalities to express your idea.
Use natural language to explain the motion, camera, style, and references you want to use.
Generate your video, then extend, edit, or refine it through targeted adjustments.
See how Seedance 2.0 is transforming modern creative workflows.
Seedance 2.0’s multi-modal input is a game-changer. I can finally reference a dance video and apply it to any character I want.
The reference capability is mind-blowing. It perfectly replicated the camera movement and pacing from my film clip.
Finally, character consistency that actually works. Faces, clothing, and even small text remain stable throughout.
The video extension feature is seamless. It feels like having an AI editor that genuinely understands continuity.
Referencing trending templates has increased our content output dramatically while keeping the quality consistent.
All plans include access to core generation features and watermark-free exports.
800 credits/month
1,600 credits/month
4,000 credits/month
10,000 credits/month
The studio includes image workflows for Seedream and Nano Banana Pro, plus video workflows for Seedance, Veo, Sora, Wan, and Kling models. Availability can vary by provider.
Seedance 2.0 is a multi-modal AI video generation experience supporting image, video, audio, and text inputs. It lets you reference motion, effects, camera movements, characters, scenes, and sound through natural language.
You can combine up to 9 images, 3 reference videos, 3 audio files, and a detailed natural-language prompt in one creative workflow.
Yes. Upload a reference clip and describe which movement, pacing, transition, or choreography you want to transfer into the new video.
The interface supports durations from 4 to 15 seconds, common cinematic aspect ratios, and output resolutions from 480p through 4K.
Generated videos can be downloaded without a watermark and are ready for use in your creative projects.
Choose a mode, upload your references, describe your vision, select the output settings, and press Generate.
Reference anything, edit anything, and create cinematic content with natural language control.