Create with Seedance 2.5, then compare available Seedance AI models for text, image, video, and audio workflows.
Click to upload assets
Max 50 files (img ā¤30, vid ā¤10, audio ā¤10)
Seedance AI is a family of video generation models for turning written prompts, still images, and supported reference media into finished clips. The same workspace lets you move from a simple text-to-video idea to a more directed production with visual, motion, and audio references.
Seedance is not a single fixed model. Each version has a distinct creation setup, from focused prompt and frame controls to larger multimodal reference collections. That makes the version choice part of the creative brief: consider how much source material you have, whether audio is a reference or a generated output, how long the clip should be, and which resolution options your delivery needs.
For a quick decision, start with 2.5 for the newest, longer, and reference-rich workflow. Use 2.0 for complete multimodal production, Mini for rapid short-form iteration, 1.5 Pro for generated audio with text or frame guidance, and 1.0 Pro for basic silent text-to-video or image-to-video creation. The controls shown below reflect the current generators.
Start with the workflow you need, then compare inputs, reference support, audio behavior, and output options. The table reflects the controls currently available in each generator.
| Compare by | Seedance 2.5Latest model | Seedance 2.0 | Seedance 2.0 Mini | Seedance 1.5 Pro | Seedance 1.0 Pro |
|---|---|---|---|---|---|
| Best fit | Longer, reference-heavy productions | Cinematic multimodal production | Fast short-form drafts | Audio-ready text and image workflows | Straightforward silent clips |
| Input methods | Text, images, video, and audio | Text, images, video, and audio | Text, images, video, and audio | Text, start frame, and optional end frame | Text, start frame, and optional end frame |
| Reference media | Up to 50 files: 30 images, 10 videos, and 10 audio files | Up to 12 files: 9 images, 3 videos, and 3 audio files | Up to 12 files: 9 images, 3 videos, and 3 audio files | Up to two guiding images | Up to two guiding images |
| Audio | Audio references supported | Audio references supported | Audio references supported | Generated audio can be enabled | Video output without generated audio |
| Output range | 4 to 30 seconds; 480p, 720p, or 1080p | 4 to 15 seconds; 480p to 4K depending on mode | 4 to 15 seconds; 480p or 720p | 5 or 10 seconds; 480p, 720p, or 1080p | 5 or 10 seconds; 480p, 720p, or 1080p |
| Key advantage | Largest reference set and longest current workflow | Full workflow with broader resolution options | Simpler output range for quick iteration | Native audio option with frame guidance | Simple prompt or image-led generation |
Swipe horizontally to compare every model.
Each model has its own dedicated creation page, so you can open the right generator without losing the family-level comparison.
Seedance 2.5 is the latest model available here. It combines prompt-led creation with a larger pool of image, video, and audio references for directed visual stories.
Choose it when a project needs more source material, a longer continuous clip, or tighter control over characters, products, motion, and sound cues.
Seedance 2.0 guides a clip with text plus image, video, or audio material. It offers more resolution choices than the Mini edition.
Choose it for multimodal work within a 15-second window, especially when resolution flexibility matters. Standard and fast modes offer different output ranges.
Seedance 2.0 Mini keeps multimodal references in a focused generator for rapid concept passes, social drafts, and motion tests at 480p or 720p.
Choose Mini to test framing, pacing, or reference combinations before committing to a larger workflow. Its compact output range keeps early decisions simple.
Seedance 1.5 Pro uses a focused text or image workflow. Guide motion with a start and optional end frame, then choose whether to generate audio.
Choose it when generated sound is central to a short clip and you do not need the larger reference library of the 2.x models.
Seedance 1.0 Pro turns a written prompt or still image into a silent clip, with compact controls for motion studies, simple concepts, and image animation.
Choose it for a direct text or image workflow when sound will be handled separately. Start and optional end frames provide basic visual guidance.
Practical answers based on the model controls currently available here.
It is a family of video generation models with different combinations of prompts, images, video references, audio, duration, and resolution. Compare those workflows here before opening a dedicated model generator.
Seedance 2.5 is the latest option shown here. Its generator supports text-only and reference-based creation, clips from 4 to 30 seconds, 480p to 1080p output, and the largest reference pool among the listed versions.
Use 2.5 for longer or reference-heavy work, 2.0 for a complete multimodal workflow with broader resolution choices, and 2.0 Mini for quick short-form drafts. Choose 1.5 Pro when generated audio and frame guidance matter. Choose 1.0 Pro for a simpler silent text-to-video or image-to-video workflow.
Yes. Every listed model supports a prompt-led workflow and an image-based option. The 2.5, 2.0, and 2.0 Mini generators also accept supported video and audio references. File counts, duration choices, and output settings vary by model.
Seedance 2.5 supports clips up to 30 seconds and a reference pool of up to 50 files, with 480p, 720p, and 1080p output. Seedance 2.0 supports clips up to 15 seconds and up to 12 reference files, while standard mode offers resolution choices through 4K. The better option depends on whether duration, reference capacity, or resolution flexibility matters most.
Yes. Each model in this comparison links to a dedicated generator. Older versions remain useful when their controls fit the job, such as Mini for compact drafts, 1.5 Pro for generated audio, or 1.0 Pro for silent text and image workflows.