Turn text or a reference image into cinematic AI video with controllable duration, resolution, aspect ratio, and native audio where supported.
Model names, settings, and credit estimates are shown before every generation.
Explore community videos created with Cdance AI. Browse examples, inspect prompts, and reproduce the workflow with one click.
Each path explains the live models, compatible controls, and practical preparation before you spend credits.
Direct a 4–30 second shot from a written scene brief, depending on the selected model.
Open workflow →
Animate a source image with model-specific duration, resolution, camera, and sound controls.
Open workflow →
Start from original templates for products, social video, people, food, architecture, and nature.
Open workflow →
Cdance AI is designed as a creator workflow, not a single-model demo. Use prompts, reference assets, motion controls, and sound options in one place.
Combine prompts with reference images, motion videos, and audio inputs to steer style, framing, and movement more precisely.
Keep the same character identity across multiple shots and scenes for stories, mascots, serialized content, and campaigns.
Generate ready-to-share videos with environmental sound, audio-driven motion, or dialogue-aware workflows where supported.
Extend clips, preserve context, and keep visual continuity when you need a longer sequence instead of a one-off sample.
Guide movement with prompt intent and reference motion so your shots feel directed instead of randomly animated.
Move from idea to output quickly with a unified editor, reusable inputs, and multiple generation modes.
See the cost before every generation. Failed provider tasks return generation credits automatically.
Credits
500 credits monthly
About 5 five-second Cdance text-to-video generations at the current standard rate
Features
Credits
1,200 credits monthly
About 12 five-second Cdance text-to-video generations at the current standard rate
Features
Credits
3,000 credits monthly
About 31 five-second Cdance text-to-video generations at the current standard rate
Features
Generation counts are estimates, not quotas. Actual credit use depends on the selected model, duration, resolution, and optional features. The editor displays the exact estimate before you submit.
New users receive free credits upon registration. Once credits are used up, you can purchase a subscription plan to continue generating videos.
We support both text-to-video and image-to-video modes. You can describe a scene with text, or upload a reference image and animate it with a prompt.
Duration is model-specific. Current video models offer choices between 3 and 30 seconds; the editor only shows values supported by your selected model.
Each video generation costs credits based on duration (base cost + per-second cost). Your credit balance is shown in the top navigation bar. Subscribe to a plan to receive monthly credits.
Yes. Once a video is generated, you can preview it directly on the page and download it. All your generated videos are also saved in the "My Videos" dashboard.
The current catalog includes Cdance Video Generator, Seedance 2.5, Seedance 2.0, Seedance 1.5 Pro, Kling 3.0 Pro, WAN 2.6 and 2.5, motion-control models, and multiple image models. The Models page and editor are the source of truth.