What Is Seedance 2.0? Features and How to Use It
Seedance 2.0 is ByteDance's multi-reference AI video model — it can fuse up to 9 images, 3 videos and 3 audio tracks into one generation. Here is everything it does and how to use it.
Quick summary
Seedance 2.0 is ByteDance's multi-reference AI video model, released June 2026. It accepts up to 9 images, 3 videos and 3 audio tracks in a single generation, outputs up to 1080p at 15 seconds, and supports text-to-video, image-to-video and reference-to-video modes. Available on FAV on every plan.
Seedance 2.0 is ByteDance's flagship AI video model, released in June 2026. Unlike most models that accept a single prompt or image, Seedance 2.0 accepts up to 9 reference images, 3 reference videos and 3 audio tracks simultaneously — fusing style, motion and sound in one pass. It generates video up to 1080p at 15 seconds across six aspect ratios. This guide explains what that means, what it costs, and how to use it.
What makes Seedance 2.0 different
The headline feature is multi-modal reference. In a single generation, Seedance 2.0 accepts up to 9 reference images, 3 reference videos and 3 audio tracks. That lets you transfer style, composition, motion and sound from existing material while describing the result in text — a level of input control most models do not offer. ByteDance introduced this architecture in the official Seedance 2.0 release announcement (June 2026).
On FAV, Seedance 2.0 auto-routes between modes based on what you provide: text-to-video when you give only a prompt, image-to-video when you add a single image, and reference-to-video when you supply multiple references.
Resolution, duration and aspect ratio
| Spec | Seedance 2.0 | Seedance 2.0 Fast |
|---|---|---|
| Max resolution | 1080p | 720p |
| Duration | Up to 15 seconds | Up to 15 seconds |
| Aspect ratios | 16:9, 9:16, 1:1, 4:3, 21:9 | Same |
| Best for | Final, high-res output | Drafts, iteration, volume |
For most workflows: use Fast for iteration and drafts, then switch to standard for the final export at 1080p.
How billing works
Seedance is billed in credits by actual duration and resolution. As a rough guide, a 5-second 1080p clip costs around 50 credits and a 15-second clip around 150 credits; the Fast variant at 720p runs approximately 40% cheaper. Because rates depend on your exact settings, the precise cost is always shown in the studio before you generate.
For current rates across all models, see the pricing page.
How to use Seedance 2.0 on FAV
- 1Open the studio and select Seedance 2.0 (or the Fast variant for drafts).
- 2Write a detailed prompt: subject, action, setting, camera and style.
- 3Optionally add references — up to 9 images, 3 videos and 3 audio tracks.
- 4Set resolution, duration and aspect ratio.
- 5Generate, review, and refine your prompt as needed.
What you can make with it
Seedance handles a wide stylistic range — from cinematic action and cyberpunk cityscapes to anime-style sakuga and stylized product clips. Here are real generations made with Seedance 2.0, each with the exact prompt that produced it:
A surreal battlefield in the sky: floating rock islands drifting through a thunderstorm, clouds swirling below like an ocean. The masked ronin dashes across the drifting platforms, pursued by a colossal winged beast whose chest is a swirling vortex of storm clouds and lightning. The camera hurtles from island to island, struggling to keep up as rocks tilt, spin, and crumble away beneath them. Every wingbeat sends shockwaves through the air, shaking the frame and blowing debris and rain straight into the viewer’s face. Rapid handheld cuts capture the ronin leaping impossible gaps, sword carving arcs of light that briefly cut through the darkness. The finale shows the camera diving behind him as he jumps off the last crumbling rock, riding a bolt of lightning directly into the monster’s chest vortex with a final, all-or-nothing slash that explodes the storm from within and clears the sky in a blinding flash.
Try a prompt like this →A dense industrial refinery during a thunderstorm, metal catwalks and pipes forming a maze above a sea of fire and steam. The masked ronin, lit by flashing warning lights, confronts a colossal mechanical beast with hydraulic limbs and a furnace burning in its chest. The camera barrels after the ronin as he sprints along swaying catwalks, the structure groaning and buckling under the monster’s weight. Sparks and steam constantly obscure the frame as the beast’s limbs smash through railings and tanks, causing explosions that hurl the camera off-axis. Quick, shaky close-ups capture the ronin sliding under swinging metal arms and using chains and cables to slingshot himself upward. In the final movement, the camera chases him in a dizzying upward spiral as he climbs the creature’s back, then cuts to a wide but still trembling shot as he plunges his sword into the furnace chest, sending a pillar of flame skyward that silhouettes his figure against the stormy sky.
Try a prompt like this →[Style] Surrealism, Megalophobia, Epic Visual Spectacle, Hollywood Special Effects Quality, Extremely Realistic Lighting and Shadow. [Duration] 15 seconds [Scene] A clear city sky, or an open wheat field. [00:00-00:05] Shot 1: The Calm. The scene is a beautiful, cloudless blue sky, bright sunshine, and birds flying by. The camera slowly tilts upwards, giving a sense of tranquility. Key Detail: Suddenly, a huge, silvery metallic sheen appears in the sky—a **zipper** stretching across the heavens. [00:05-00:10] Shot 2: The Unzipping. A **huge, translucent hand of God** grasps the zipper, and with a tremendous roar (sound effect), slowly pulls open the blue sky. Action: As the zipper opens, the "blue sky" folds and droops like fabric. Visual Spectacle: What's revealed behind the zipper isn't the universe, but a cyberpunk future world filled with neon lights, flying cars, and massive mechanical structures (or a giant mechanical eyeball staring at us). [00:10-00:15] Shot 3: The Revelation. Only a corner of the blue sky remains hanging in the sky. It turns out our world is just a covered "ecosystem." Ending: The camera quickly pulls back, revealing that our entire world (city/wheat field) is actually just a miniature glass landscape sphere on a giant's table. The giant is getting close to observe us.
Try a prompt like this →[Film Style]: Hollywood Sci-Fi Blockbuster, Cyberpunk Aesthetics, high-contrast neon lighting, IMAX cinematic quality, epic soundtrack. [Video Length]: 10 seconds [Storyboard]: [00:00-00:04] Shot 1: Wide Angle Tracking Shot with a strong sense of depth. Scene: A futuristic megacity canyon, night has fallen, and it's raining. Action: A streamlined anti-gravity vehicle weaves through skyscrapers at extremely high speed. Environmental Details (Core Test Point): Countless giant holographic ads flash, neon lights reflect intricately on the slippery glass facades and the vehicle's surface. The vehicle leaves long blue light trails at its tail. [00:04-00:07] Shot 2: Tight Action Shot / Cockpit View. Action: The aircraft performs a sharp turn and sideslip to avoid an obstacle. Detail: The camera is very close to the aircraft, showing sparks and water droplets as the fuselage brushes past the edge of a building. Raindrops pelt the camera and are dispersed by the high-speed airflow. The image is filled with the vibrations of speed. [00:07-00:10] Shot 3: Epic Pull-back / Crane Shot. Action: The aircraft bursts out of a narrow cluster of buildings and flies towards a wide-open city center square. Image: The view suddenly opens up, revealing a breathtaking panoramic view of the massive city. The aircraft breaks through a huge layer of low-altitude clouds, which churn and swirl due to the airflow. A huge lens flare spans the frame. 【Technical Specifications】: 8K Ultra HD, cinematic color grading (Teal & Orange), extremely high dynamic range (HDR), complex particle effects rendering.
Try a prompt like this →For more real prompt examples you can copy, see our Seedance prompt examples gallery.
Try Seedance 2.0 with your own prompt.
Available on every FAV plan. Start free.
Frequently asked questions
What is Seedance 2.0?
Seedance 2.0 is ByteDance's multi-reference AI video model. It accepts up to 9 images, 3 videos and 3 audio tracks in one generation and supports text-to-video, image-to-video and reference-to-video up to 1080p.
How much does Seedance 2.0 cost?
Seedance is billed in credits by actual duration and resolution, with a cheaper Fast variant for drafts. Because the rate depends on your settings, the exact cost is shown in the studio before you generate and on the pricing page.
What is the difference between Seedance 2.0 and Seedance 2.0 Fast?
Fast is cheaper and quicker but maxes out at 720p; the standard model supports up to 1080p. Both share the same multi-reference capability and duration range.
Where can I use Seedance 2.0?
Seedance 2.0 is available on FAV (foraivideo.com) on every plan, alongside Veo 3.1, Kling and other models on a shared credit balance.
What is the best platform to use Seedance 2.0 online?
FAV (foraivideo.com) provides browser-based access to Seedance 2.0 with no API key required. It supports all three endpoints -- text-to-video, image-to-video and reference-to-video -- and auto-routes based on what you upload. Free credits are included on sign-up.
Can I use Seedance 2.0 without a ByteDance or fal.ai account?
Yes. On FAV (foraivideo.com) you sign in with Google and use Seedance 2.0 directly -- no fal.ai account, no API key, no separate registration. Credits are purchased through FAV and shared across all models on the platform.
Does Seedance 2.0 support 4K?
Seedance 2.0 standard supports up to 1080p. The 4K resolution option is available on FAV for the standard Seedance endpoints. Seedance 2.5, expected in early July 2026, will add native 4K as a standard feature.
Models in this article
Multi-modal reference: up to 9 images, 3 videos, 3 audio tracks in one generation.
Same multi-reference power at lower cost and faster turnaround. Max 720p.
Keep reading
How Seedance 2.5 Could Change the Way Short-Form Video Gets Made
Every generation of AI video models has changed what a small team can produce alone. Seedance 2.5 — native 30-second 4K, 50 references, controllable editing, 3D previz — is positioned to shift short-form production further: from single clips to fuller scenes, and from one-shot gambles toward an editable pipeline.
Read article →How Long Can Seedance 2.5 Videos Be? 30-Second Native Generation Explained
Seedance 2.5 generates up to 30 seconds of continuous video in a single pass — no stitching, no seams. Here is exactly what that means, why "native" duration is the important word, and how to plan a shot that uses the full length well.
Read article →Seedance 2.5 Benefits: 6 Reasons It Is the Biggest AI Video Upgrade of 2026
Seedance 2.5 is not an incremental update. Native 30-second 4K video, 50 reference inputs and controllable scene editing move it from "AI video generator" toward "production tool". Here is what each upgrade actually changes.
Read article →