Page loading...
Page loading...
Generate cinematic videos of 4โ12 seconds with up to 1080p resolution, and keep dialogue, sound effects, and visuals synchronized natively.
Seedance 1.5 Pro is ByteDance's Seed model for high-quality audiovisual creation, putting high-fidelity visuals and native synchronized audio into one generation. Raphael supports text-to-video, first-frame, and first-and-last-frame image-to-video, with 480p, 720p, or 1080p options, and generates 4โ12 second videos. It is suitable for character dialogue, product demos, brand short films, and narrative shots that require clear panning, tilting, zooming, or rotating camera movement.
View sample outputs for music festival selfies, low-altitude canyon flight, close-ups of waves and wildlife.
Generate a 7-second, 16:9 nighttime outdoor music festival shot: a young woman holds a wide-angle camera for a selfie-style shot while smiling and moving with the crowd in rhythm; the audience cheers behind her while stage lights flicker in the distance. Keep the face sharp, the handheld tracking natural, crowd continuity consistent, and add live music and cheering sounds.
Create more production-ready audiovisual content with up to 1080p visuals, native audio, and clearly defined camera movement.

โInside a rainy-night bus stop, a short-haired female actor speaks naturally toward the camera while one hand lightly touches raindrops on the glass. Keep lip movement, dialogue, raindrops, and distant traffic ambience naturally synchronized, with realistic cinematic photography and no subtitles.โ
Audio is not attached later after the video is rendered; it is generated together with the visuals. You can specify character lines, speaking tone, motion sound effects, music, and spatial ambience in the prompt so lip sync and sound events align more naturally with the image.

โAt dawn by a window, a character slowly turns a precision mechanical wristwatch. The close-up should preserve natural skin texture, brushed metal, dial structure, glass reflections, and fabric fibers, with soft window light and shallow depth of field; keep the frame free of brand text.โ
Supports 480p, 720p, and 1080p, and allows any whole-second duration from 4 to 12. You can test lip sync and camera movement in lower resolution first, then choose 1080p for selected versions to get clearer skin, materials, lighting, and spatial details.

โLet a white paper crane on a wooden table in the first frame naturally take off, and gradually transform into a real white bird by the end frame. Keep subject continuity stable, maintain steady framing and warm indoor lighting, and transition smooth paper folds into real feathers.โ
Upload one image to let a product, character, or concept evolve naturally into motion; upload two images to constrain both the start and end of the video. The prompt fills in middle motion, camera path, and subject features that need to remain consistent.

โA deep red sports car is parked in a wet underground garage. The camera circles around the body from a low angle and pushes forward, using light strips and floor reflections to guide movement, ending at a clear product reveal on the front of the car with sharp, crisp body detail.โ
You can explicitly describe panning, tilting, dolly-in, dolly-out, orbiting, rotating, and tracking so camera motion serves the story. Combined with actor action, shot size, and timeline order, this enables product reveals, performances, and action scenes.
Switch between top video models in one generator.
Choose the right Seedance video model for image quality, reference input, audio, and cost.
| Feature | Seedance 1.0 Pro Fast | Seedance 1.5 Pro | Seedance 2.0 Mini | Seedance 2.0 |
|---|---|---|---|---|
| Resolution | 480p / 720p / 1080p | 480p / 720p / 1080p | 480p / 720p | 480p / 720p / 1080p |
| Duration | 2โ12 seconds | 4โ12 seconds | 4โ15 seconds | 4โ15 seconds |
| Reference input | Text or 1 first frame | Text or up to 2 images | Image / video / audio | Image / video / audio |
| Native audio | No | Yes | Yes | Yes |
| Credits per second | 4/8/17 | No audio 5/11/23 | 18/39 | 36/77/191 |
| Best for | Fast low-cost drafts | High-definition audiovisual scenes | Cost-efficient multimodal validation | Premium multimodal production |
Try other Seedance models and choose the right option based on quality, audio, reference support, and budget.
Free images include a watermark. Upgrade for clean outputs, faster generation, and commercial use.
Perfect for getting started
Learn about native audio, first-and-last frames, image quality, duration, and credits with or without sound.
It combines video generation at up to 1080p with native synchronized audio, making it ideal for complete audiovisual shots that require dialogue, sound effects, ambience, and cinematic camera movement.
It supports any integer duration from 4 to 12 seconds and options of 480p, 720p, or 1080p.
In silent mode, it is 5 credits per second at 480p, 11 credits per second at 720p, and 23 credits per second at 1080p.
With native audio enabled, it is 10 credits per second at 480p, 21 credits per second at 720p, and 46 credits per second at 1080p.
Without audio: 92 credits. With audio: 184 credits. The generator will show the actual total credits based on resolution, duration, and audio settings.
Yes. Text-to-video is suitable for creating scenes from scratch; if you need to retain specific characters, products, or composition, upload reference images.
A maximum of 2 images can be uploaded. One image is used for first-frame generation, and two can be used to control both the start and end of the video.
In the prompt, clearly define the speaker, full lines, tone, shot size, and ambient sound. Avoid too long lines from multiple people at once, and keep the face clearly visible in the frame.
It is ideal for character dialogue, brand short films, product demos, action shots, and mood-driven storytelling where high-definition visuals and synchronized sound are both essential.