Seedance 2.5 Flash AI Video Generator

Turn a written scene, a product image or a set of creative references into video with ByteDance's Seedance 2.5 Flash. Create 4–30 second clips in 480p or 720p, guide the opening and ending with image inputs, and choose whether to generate sound. Start creating directly in the generator below.

Video Generator

0/1536
Generation requires 90 credits

Please enter a prompt

Video Generation

Your generated videos will appear here.

What is Seedance 2.5 Flash?

Seedance 2.5 Flash is a video generation model in ByteDance's Seedance family. On Yolly AI, it brings text-to-video, image-to-video and multimodal reference generation into one workspace. You can begin with a scene description, animate a first frame, or supply visual and audio references for a shot. Resolution, clip length, aspect ratio and audio settings let you shape the output for the project you are working on.

Create with text, frames and multimodal references

Text-to-video from a scene description

Describe the subject, action, setting and camera movement you want to see. Seedance 2.5 Flash turns your prompt into a video clip without requiring an image upload. Use the same scene idea to explore different framing, lighting or actions, then refine the prompt around the version that suits your project.

Image-to-video with first and last frames

Upload an image to establish the opening frame of your video. Add a last-frame image when the ending matters to the sequence you are building. This gives product shots, illustrated scenes and visual transitions a defined starting point. Describe the motion you want between the supplied frames in your prompt.

Image, video and audio references in one workflow

In Reference mode, combine up to 10 images, 5 video clips and 5 audio clips to guide a generation. Explain the role of each reference, such as the product appearance, movement or musical direction. Reference videos and reference audio can each total up to 30 seconds per request.

Optional generated sound

Use Generate Audio when you want sound created with the picture. Describe the ambience, effects or musical mood that belongs in the scene. You can also turn the setting off when you need a silent clip for an edit or a separate soundtrack. Audio references provide another way to guide the generation.

480p and 720p video in the format you need

Choose an output duration from 4 to 30 seconds and select 480p or 720p. Landscape, portrait, square and wider aspect ratios let you plan a shot for different layouts. Image-to-video uses the supplied frame dimensions to determine the shape of the output, so prepare your input image for the intended composition.

A visible credit estimate before generation

Adjust your settings and review the required credits in the generator before submitting. The rate depends on resolution and whether you include reference video. For reference-video requests, the calculation includes both the uploaded video duration and the output duration. Images and audio references add no billable seconds.

How to use Seedance 2.5 Flash on Yolly AI

  1. Choose a starting point

    Select Text to Video, Image to Video or Reference. For an image-based shot, upload a first frame and optionally a last frame. For Reference mode, add the materials you want the model to use.

  2. Describe the shot

    Write what happens, how the camera moves and how the scene is lit. If you uploaded references, explain what to take from each one. Add sound directions when you enable audio generation.

  3. Set the output and check credits

    Choose 480p or 720p, a duration between 4 and 30 seconds, and an aspect ratio where available. Review the displayed credit estimate before clicking Generate.

  4. Review and download your video

    Follow the task's progress in the results panel. When it finishes, preview the clip and download it. Use the result to decide which motion, composition or prompt detail to refine in your next version.

Create a Seedance 2.5 Flash video

Video ideas to build with Seedance 2.5 Flash

Product and campaign concepts

Animate a product image or use references to explore a campaign scene. Test how an object moves through the frame, how the background looks and where a shot begins and ends before developing the full edit.

Social and vertical video

Build a short scene around one clear action, then choose portrait, square or landscape framing for your layout. Optional audio helps you explore a scene with ambience or effects alongside the picture.

Storyboards and motion studies

Turn a storyboard frame or written shot description into a moving draft. Explore pacing, camera directions and transitions, then use what you learn to revise the next shot or the overall sequence.

Music-led visual ideas

Add an audio reference and describe the setting, subject and visual mood you have in mind. Use the resulting clip to explore a short musical sequence, a visualizer concept or the atmosphere of a larger video.

Seedance 2.5 Flash settings and input limits

Seedance 2.5 Flash settings and input limits
SettingAvailable on Yolly AI
Generation modesText to video, image to video, multimodal reference
Output resolution480p or 720p
Output duration4–30 seconds
Aspect ratios16:9, 9:16, 1:1, 4:3, 3:4, 21:9 or adaptive
Frame inputsOne first frame, with an optional last frame
Reference filesUp to 10 images, 5 videos and 5 audio clips
Reference video duration2–30 seconds per clip, up to 30 seconds in total
Reference audio duration2–30 seconds per clip, up to 30 seconds in total
Audio outputOptional generated audio

How Seedance 2.5 Flash credits are calculated

Use the rates below to plan a generation. A request without reference video is charged for the output duration. A request with reference video is charged for the combined duration of all reference videos and the output.

How Seedance 2.5 Flash credits are calculated
ResolutionWithout reference videoWith reference video
480p18 credits per second12 credits per second
720p40 credits per second24 credits per second

For example, a 10-second 720p text-to-video clip costs 400 credits. With a 5-second reference video and a 10-second output, the estimate is (5 + 10) × 24 = 360 credits.

Reference images and audio add no billable seconds. The final credit amount is rounded up to a whole credit. The generator displays the estimate for your current inputs and settings.

Seedance 2.5 Flash frequently asked questions

1

What can I create with Seedance 2.5 Flash?

You can create video clips from written descriptions, animate images and guide scenes with image, video or audio references. Product concepts, short social scenes and storyboard motion studies are useful starting points for exploring the model.

2

Which input mode should I choose?

Start with Text to Video for a written scene idea. Use Image to Video when you have a specific opening image. Choose Reference when you want uploaded pictures, video clips or audio to guide the subject, movement or mood.

3

How do first and last frames work?

The first frame establishes the opening image. You can add a last frame to specify the ending and describe the action between them. A last frame requires a first frame. This image-input mode is used separately from multimodal references.

4

How should I describe multiple reference files?

Give each reference a clear purpose in your prompt. For example, describe which image supplies the product appearance, which video guides the movement and which audio sets the mood. Keeping these instructions specific makes the intended shot easier to communicate.

5

Can I use an audio reference without images or video?

Yes. In Reference mode, you can add an audio clip and a prompt describing the visuals you want to create. Explain the scene and how it should relate to the sound, then choose your output settings.

6

How do I choose between 480p and 720p?

480p is a useful setting for reviewing movement and scene ideas at a lower credit cost. 720p provides a higher output resolution for reviewing visual detail. The current resolution and duration determine the credit estimate shown in the generator.

7

Can I make portrait or square videos?

Yes. Text and reference generation include 9:16 portrait and 1:1 square options alongside landscape and wider ratios. For image-to-video, prepare the first-frame image in the composition you want because the aspect ratio follows the supplied frame.

8

Does every video include generated audio?

You control that with Generate Audio. Enable it when the scene needs generated sound, or switch it off for a silent video. An uploaded audio reference is an input that guides the scene, while the audio setting controls generated sound.

9

Why does a reference video affect the credit estimate?

Reference-video requests use their own per-second rate and include the duration of the input videos as well as the requested output. The total duration of reference audio or images does not add billable video seconds.

10

Where do I find the finished video?

Your task appears in the results panel while it is being processed. Once it completes, you can preview and download the video there. The history control lets you return to previous generations associated with your account.

Seedance 2.5 Flash AI Video Generator | Yolly AI