HushgenUse Seedance 2.5 on Hushgen to turn a text prompt, an image or an existing clip into a 4–30 second video with generated audio, at 480p up to 4K.
Official Seedance 2.5 clips in different lengths and aspect ratios. Hover to preview, select to watch with sound.
Seedance 2.5 is a video generation model developed by ByteDance and released on July 31, 2026. It generates a single continuous clip of up to 30 seconds, creates the audio together with the picture, and accepts up to 30 images, 10 video clips and 10 audio clips as references in one request.
On Hushgen, Seedance 2.5 is available in three tools: Text to Video (a prompt, optionally with reference files), Image to Video (a first frame, optionally with a last frame) and Video Extend (continue an existing clip). You choose the duration, resolution and whether to generate audio, and the credit cost is shown before you generate.
You do not need a ByteDance account to use Seedance 2.5 here. Hushgen credits work with Seedance 2.5 and with every other model on the site, including Kling 4.0, Veo 3.1 and Wan 2.7, so one plan covers several video models.
“Seedance 2.5 can generate high-quality, 30-second audio-video clips in a single pass.”
“up to 30 images, 10 video clips, and 10 audio clips as reference materials in a single pass.”
Hushgen is an independent platform that provides Seedance 2.5 through licensed API services. Hushgen is not affiliated with or endorsed by ByteDance.
What changes compared with earlier Seedance models, shown with official sample footage.
Official Seedance 2.5 sample
Seedance 2.5 doubles the 15-second limit of Seedance 2.0. A reveal, a walk through a crowd or a full product demo can play out in a single shot, without stitching clips together.
Official Seedance 2.5 sample
Dialogue, effects and ambience are created together with the video and stay in sync, including scenes where several characters speak.
Official Seedance 2.5 sample
In Text to Video, pick 16:9, 9:16, 4:3, 3:4, 1:1 or 21:9 before you generate. In Image to Video the clip keeps the shape of your first frame.
Official Seedance 2.5 sample
Add up to 30 images, 10 videos and 10 audio files to a Text to Video request, so the product, character, motion or voice in the result matches your material.
In Image to Video, the first image fixes how the clip starts and an optional second image fixes how it ends.
Video Extend adds 4–30 seconds to an existing clip. Repeat it on the result to build a longer sequence.
Use 480p or 720p for drafts and 1080p or 4K for final delivery. Higher resolutions cost more credits per second.
Seedance 2.5 runs in three Hushgen tools. Each one lists Seedance 2.5 in its model menu.
Describe a scene and Seedance 2.5 generates it. Add reference files to keep a character, product, motion or sound consistent with your material.
Open Text to Video
Animate a still photo, illustration or product shot. Add a last frame to control where the motion ends.
Open Image to Video
Continue a clip that ends too soon. Run it again on the result to build a longer sequence shot by shot.
Open Video Extend| Tool | You provide | You get |
|---|---|---|
| Text to Video | A text prompt. Optional: up to 30 reference images, 10 reference videos and 10 reference audio files. | A 4–30 second video at the resolution and aspect ratio you choose. |
| Image to Video | One image used as the first frame and a text prompt. Optional: a second image used as the last frame. | A 4–30 second video that starts from your image, in the same aspect ratio as the image. |
| Video Extend | An existing video clip and a prompt describing what happens next. | 4–30 seconds of new footage that continues the clip. |
Use the generator on this page, or open Text to Video, Image to Video or Video Extend. Seedance 2.5 is the default model in all three.
Write a prompt. For Image to Video, upload the first frame and optionally a last frame. For Video Extend, upload the clip to continue. In Text to Video you can also add reference images, videos or audio.
Pick a duration from 4 to 30 seconds, a resolution from 480p to 4K, an aspect ratio (Text to Video only) and whether to generate audio. The credit cost updates as you change settings.
Sign in and select Generate. Credits are deducted when the task starts and returned automatically if the generation fails.
Watch the result, download the MP4, or find it later in My Creations. You can extend it or regenerate with a changed prompt.
Write prompts like a shot list: subject, action, camera, light and sound. Put spoken lines in double quotes.
Text to Video
30s · 1080p · 16:9 · audio on
A 30-second commercial for a matte black wireless speaker, one continuous shot. 0–8s: slow dolly in across a dark wooden desk at night, warm lamp light. 8–18s: the camera circles the speaker as it lights up and music starts. 18–30s: pull back to reveal a cozy living room, soft rain on the window. Sound: lo-fi music from the speaker, light rain.
Text to Video
15s · 720p · 16:9 · audio on
A rainy night inside a small Tokyo noodle bar. An older chef wipes the counter and says, "You came back." A young woman in a wet coat sits down and replies, "I never left." Medium shots, warm neon light through steamed windows, quiet rain and kitchen sounds.
Text to Video
10s · 1080p · 9:16 · audio on
Handheld selfie-style vertical video of a hiker reaching a snowy mountain ridge at sunrise. She turns the camera to reveal the valley below, wind in her hair, breath visible in the cold air. Natural wind sound and footsteps on snow.
Image to Video
First frame: your photo · 8s · 1080p
The camera slowly orbits the product while soft studio light sweeps across its surface. Small particles drift in the background. Keep the product shape, color and label exactly as in the image.
Seedance 2.5 is billed in Hushgen credits per second of video. Credits are shared with every model and come with plans and credit packs.
| Resolution | Per second | 5-second video | 10-second video | 30-second video |
|---|---|---|---|---|
| 480p | 72 credits | 360 credits | 720 credits | 2,160 credits |
| 720p | 144 credits | 720 credits | 1,440 credits | 4,320 credits |
| 1080p | 360 credits | 1,800 credits | 3,600 credits | 10,800 credits |
| 4K | 720 credits | 3,600 credits | 7,200 credits | 21,600 credits |
Both models are available on Hushgen. Choose by clip length, reference needs and budget.
| Feature | Seedance 2.5 | Seedance 2.0 |
|---|---|---|
| Duration per generation | 4–30 seconds | 4–15 seconds |
| References in Text to Video | 30 images, 10 videos, 10 audio | 9 images, 3 videos, 3 audio |
| Reference video / audio length | Up to 30 seconds each | Up to 15 seconds each |
| Resolution on Hushgen | 480p – 4K | 480p – 4K |
| Image to Video aspect ratio | Follows the image | Selectable |
| Dedicated Reference to Video tool | No (references are added in Text to Video) | Yes |
| Video Extend | 4–30 seconds, up to 4K | 4–15 seconds, up to 1080p |
| Credits per second at 720p | 144 | 96 |
Choose Seedance 2.5 when one take must run longer than 15 seconds, or when you need more than 9 reference images.
Choose Seedance 2.0 for shorter clips at a lower price, or when you want to set the aspect ratio in Image to Video.
What Seedance 2.5 on Hushgen does not do, so you can decide whether it fits your project.
Seedance 2.5 is a video generation model from ByteDance, released on July 31, 2026. It makes clips of up to 30 seconds with audio generated together with the picture, and accepts up to 30 images, 10 videos and 10 audio files as references.
Yes. Seedance 2.5 is available in Hushgen’s Text to Video, Image to Video and Video Extend tools. You need a Hushgen account and credits to generate.
No. Hushgen is an independent platform that provides Seedance 2.5 through licensed API services. It is not affiliated with or endorsed by ByteDance.
A text prompt in every tool. Image to Video also needs one image as the first frame and accepts an optional last frame. Video Extend needs the video clip to continue. Text to Video can take optional reference images, videos and audio.
An MP4 video of 4 to 30 seconds at 480p, 720p, 1080p or 4K, with generated audio unless you turn audio off.
Each generation is 4 to 30 seconds. To make a longer video, use Video Extend on the result to add further segments.
Yes. Audio is generated with the video and stays in sync with it. You can turn it off in the tool; the price is the same either way.
In Text to Video, up to 30 images, 10 videos and 10 audio files per request. Reference videos and reference audio can each total up to 30 seconds.
Seedance 2.5 is billed in credits per second of video. A 5-second 720p clip costs 720 credits and a 30-second 4K clip costs 21,600 credits. The pricing table on this page lists other combinations, and the tool shows the exact cost before you generate.
Seedance 2.5 makes clips up to 30 seconds instead of 15 and accepts more references (30 images, 10 videos, 10 audio instead of 9, 3 and 3). Seedance 2.0 costs fewer credits per second and lets you choose the aspect ratio in Image to Video.
A clear, well-lit image in the aspect ratio you want the video to have, with the main subject fully in frame. The video keeps the image’s aspect ratio.
Yes. Add parameters to this page’s URL to pre-fill the generator, for example /seedance-2-5?model=bytedance/seedance-2.5/text-to-video&duration=10&quality=1080p&prompt=A%20red%20fox%20in%20snow. Supported parameters are model, tab, duration, quality (480p, 720p or 1080p), audio=0 to turn audio off, and prompt. The user still reviews the settings and selects Generate.
Not at the moment. Seedance 2.5 on Hushgen is used through the web app.
One credit balance works with all of them.
Video models
Models are provided by their developers through licensed API services. Hushgen is an independent platform; model names belong to their respective owners. Disclaimer
