Generate beautiful assets with Seedance 2.5
Seedance 2.5 generates coherent 30-second videos with reference control and targeted edits.
Seedance 2.5
30-second cosmic horror movie trailer, 1990s theatrical style, complete with title cards. Character A: a marine biologist in her 40s, salt-crusted parka, chapped face, deep exhaustion in her eyes. Character B: an old harbor master in his 70s, milky left eye, oilskin coat, trembling hands. Setting: a decaying fishing village on a black ocean, thick fog, crooked docks, a dead lighthouse. Structure: Cold open (0-4s): slow aerial descent through fog toward the village, the ocean unnaturally still, a vast dark shape suggested beneath the surface; low drone building. Beat one (4-9s): quick cuts, abandoned boats, nets dropped mid-haul, a meal still steaming on a crate, the biologist walking the empty dock; whispered voice-over tone. Title card (9-11s): black screen, white serif text: 'THIS SUMMER'. Beat two (11-17s): the harbor master grabs her arm and points at the water; the tide pulls back too far, too fast, exposing monolithic carved stones that should not exist; thousands of seabirds flee inland. Title card (17-19s): black screen: 'SOME THINGS SLEEP'. Beat three (19-25s): accelerating montage, the fog glowing sickly green from below, villagers running, the lighthouse flickering back to life on its own, the biologist screaming something we cannot hear, waves rising; music swelling then cutting to silence. Final card (25-28s): title reveal in weathered serif letters: 'THE DEEP BELOW', text slowly warping as if underwater. Stinger (28-30s): one last shot, the ocean surface, calm, then a colossal eye opens just beneath it; hard cut to black with a deep boom. Trailer editing grammar: rhythmic cuts synced to sound design, dread through suggestion, the creature never fully shown until the final stinger. Desaturated cold palette with sickly green undertones, heavy fog diffusion, 35mm film grain, deep sub-bass rumble, distant foghorn.
Seedance 2.5
30-second cosmic horror movie trailer, 1990s theatrical style, complete with title cards. Character A: a marine biologist in her 40s, salt-crusted parka, chapped face, deep exhaustion in her eyes. Character B: an old harbor master in his 70s, milky left eye, oilskin coat, trembling hands. Setting: a decaying fishing village on a black ocean, thick fog, crooked docks, a dead lighthouse. Structure: Cold open (0-4s): slow aerial descent through fog toward the village, the ocean unnaturally still, a vast dark shape suggested beneath the surface; low drone building. Beat one (4-9s): quick cuts, abandoned boats, nets dropped mid-haul, a meal still steaming on a crate, the biologist walking the empty dock; whispered voice-over tone. Title card (9-11s): black screen, white serif text: 'THIS SUMMER'. Beat two (11-17s): the harbor master grabs her arm and points at the water; the tide pulls back too far, too fast, exposing monolithic carved stones that should not exist; thousands of seabirds flee inland. Title card (17-19s): black screen: 'SOME THINGS SLEEP'. Beat three (19-25s): accelerating montage, the fog glowing sickly green from below, villagers running, the lighthouse flickering back to life on its own, the biologist screaming something we cannot hear, waves rising; music swelling then cutting to silence. Final card (25-28s): title reveal in weathered serif letters: 'THE DEEP BELOW', text slowly warping as if underwater. Stinger (28-30s): one last shot, the ocean surface, calm, then a colossal eye opens just beneath it; hard cut to black with a deep boom. Trailer editing grammar: rhythmic cuts synced to sound design, dread through suggestion, the creature never fully shown until the final stinger. Desaturated cold palette with sickly green undertones, heavy fog diffusion, 35mm film grain, deep sub-bass rumble, distant foghorn.



30-second long-narrative video generation
Create a full ad, product demo, or short drama in one pass, steer it with references, then fix only the region that needs work.
Single-pass narrative clips
Build a complete sequence with opening, progression, and ending in one generation, reducing visible stitch points.
50 references in one prompt
Mix images, video, and audio to control characters, products, scenes, and style.
Region-level editing
Change a product angle, background object, or character detail without regenerating the full video.
Consistent characters and scenes
Maintain identity, lighting, motion, and continuity across multi-character shots and scene changes.
Commercial video formats
Create ads, ecommerce product demos, short dramas, social clips, and film previz with controlled pacing.
Reference-to-video shot guidance
Use rough blocking, green-screen footage, or model references to guide motion, staging, and camera framing.
Create with Seedance 2.5 with full control
Create with Seedance 2.5 with full control
Select Seedance 2.5
Choose Seedance 2.5 from the ElevenLabs video model library when you need long-narrative generation with reference control.
Enter prompt and add references
Write the scene arc, then add reference images or videos for characters, products, locations, style, and camera direction.
Generate and refine
Generate the video, review continuity and details, then refine the prompt or use region-level edits for targeted fixes.
Frequently asked questions
Get started with Seedance 2.5 for free
The best image, video, and audio models — all inside ElevenCreative. Start generating today.

Discover more image & video generation models
Explore our full library of AI image and video generation models, each with unique strengths and capabilities.

Flux 3 Video generates grounded, multimodal AI video from prompts, images, and clips.

Seedream 5.0 Pro creates controllable commercial images, dense layouts, and realistic product visuals.

Wan 2.5 Video generates short, high-resolution videos from text or images with smooth motion and flexible control.

Seedream 5 Lite rapidly generates images up to 3K with deep reasoning, precise prompt-following

HeyGen Avatar 4 quickly generates expressive, lifelike talking avatars from a single photo

Creatify Aurora transforms a single photo and audio into ultra-realistic AI avatar videos with nuanced motion and visual style

Veed Fabric 1.0 creates realistic talking videos from any image, with lifelike gestures, lip sync, and flexible style control.

Sync 3 offers 4K lip synced video generation with shot-level realism, emotion preservation

Gemini Omni Flash creates realistic, high-resolution video from text, images, and video prompts all on ElevenLabs.
