How to Create AI Videos with Seedance 2.5 & 2.0

Seedance is an AI video generation model that lets you create a wide range of scenes and motion using text, image, video, and audio references.

AI Studios supports both Seedance 2.5 and Seedance 2.0. Choose the model based on your desired video length and generation needs, then use prompts and references to create your video.

Seedance 2.5 can generate videos up to 30 seconds long, making it well suited for longer scenes or videos with multiple sequential actions.

In addition to text prompts, you can use image, video, and audio references to guide characters, products, motion, camera work, visual style, and sound.

It’s especially useful for ads, UGC, story-driven content, and other videos that involve multiple actions or a continuous sequence within a single generation.

Seedance 2.0 is a multimodal AI video generation model that lets you create videos using text and various types of references.

You can generate a new scene from text, animate an image, or combine image, video, and audio references to guide characters, motion, camera work, visual style, and sound.

Both Seedance 2.5 and Seedance 2.0 support video generation with text and various types of references.

Seedance 2.5 supports videos up to 30 seconds, while Seedance 2.0 supports videos up to 15 seconds. Choose the model based on your desired video length and the generation options available for each model.

If you want to create a longer scene or a video with multiple sequential actions, Seedance 2.5 is a good option with support for generations up to 30 seconds.

From the AI Studios dashboard, go to Seedance 2.5 or Video Generator.

From the model list, select Seedance 2.5 or Seedance 2.0.

Available video lengths and generation options may vary by model.

Enter a prompt describing the scene and motion you want to create.

In addition to text, Seedance supports image, video, and audio references.

Using text only

Describe the subject, action, scene, and other details in natural language to generate a new video.

Using an image

Add an image as a reference and describe how you want the person, product, or other subject in the image to move.

Using multiple references

Combine image, video, or audio references to guide elements such as characters, motion, camera work, visual style, or sound.

Uploaded references can be specified in your prompt using tags such as @image1, @video1, and @audio1.

Before generating your video, configure your preferred generation options.

Available options may vary depending on the Seedance model you select.

Select the length of the video you want to generate.

Seedance 2.0 supports duration settings from 4 to 15 seconds.

Select the aspect ratio for your video.

Choose a ratio that fits your content and the platform where you plan to use it.

Select your preferred video resolution from the available options.

When supported, you can choose whether to generate audio with your video.

With audio enabled, Seedance can generate dialogue, background audio, music, or sound effects along with the video. Seedance 2.0 also supports an Audio On/Off option for generating video with or without audio.

Once you’ve reviewed your prompt, references, and generation settings, click Generate.

Video generation will begin, and you can review the result once it’s complete.

If the result isn’t what you expected, adjust your prompt or references and try generating again.

Seedance 2.5 and Seedance 2.0 are available to AI Studios paid plan users.

Generating a video with Seedance uses Generative Credits. The number of credits required depends on the model and generation options you select.

Before generating, check the estimated credit cost displayed on the Generate button. The required credits may vary depending on settings such as the selected model, video duration, and resolution.

If you don’t have enough Generative Credits, you can purchase additional credits to continue generating videos.

If video generation fails due to a system or generation error, the credits used for the failed generation are automatically refunded.

Generation may be restricted if your prompt or uploaded files use unsupported formats or don’t meet applicable content policies.

Q. What aspect ratios are supported?

16:9 (YouTube/Web), 9:16 (Reels/TikTok/Shorts), 1:1 (Instagram), 4:3 (Presentation)

Q. Who owns the generated video?

Ownership belongs to the user under AI Studios Terms of Service, with responsibility for not infringing third-party rights.

Q. Can I edit generated videos?

Yes. You can reuse generated videos as inputs to modify or extend scenes.

Q. How do @ mentions work?

Uploaded files are automatically tagged (e.g., @image1, @video1, @audio1).

Including these tags in your prompt tells the model how to use each asset.