AI video generator with real human faces — built on the Seedance 2.0 model, designed to help creators produce short videos with consistent, recognizable faces that most text-to-video tools struggle with.
🌐 Homepage: https://seedance2.cloud 📋 Pricing: https://seedance2.cloud/pricing 🎬 Gallery: https://seedance2.cloud/showcases
Seedance 2 Cloud is a web platform that lets indie creators, social media managers, and small marketing teams generate short AI videos for TikTok, Reels, Douyin, and YouTube Shorts — without writing code, training models, or stitching clips manually.
The platform wraps the Seedance 2.0 video model with a template library, image-reference control, and a multi-provider failover backend, so generations don't stall when one upstream is overloaded.
- Real human faces: generate videos featuring consistent, recognizable faces without the awkward artifacts common in competing text-to-video tools.
- Multi-provider redundancy: runs on a failover backend that auto-switches between providers so a single upstream outage doesn't break your workflow.
- Transparent credit pricing: pay only for what you generate. Free trial credits let you evaluate quality before committing.
- Built for creators, not engineers: template library, image-reference control, simple UI — no code required.
- Text-to-video & image-to-video: 720p and 1080p output, clip lengths from 5 seconds to 1 minute.
- Short-form social video (TikTok / Reels / Douyin / Shorts)
- Product demo clips for indie SaaS launches
- Ad creative variants for small marketing teams
- Story-driven mini-series with consistent characters across episodes
- Sign up at seedance2.cloud — free trial credits included.
- Pick a template from the gallery or start from a prompt.
- Upload a reference image (optional) for face consistency.
- Generate — typical wait time is ~5 minutes per clip.
- Download the result (MP4) or iterate on the prompt.
Full tutorials on the homepage.
Active development on:
- Multi-character templates with consistent cross-scene faces
- Voice synthesis (text-to-speech with lip sync)
- Series mode for multi-episode consistency
- Native Chinese/English bilingual captions
This repository is the public-facing profile for seedance2.cloud. The platform itself runs on Cloudflare Workers + D1 + R2, with video generation via a multi-provider backend (KIE, FAL, APIMart, aivideoapi).
For partnership or technical inquiries: hello@seedance2.cloud.
- Main site: https://seedance2.cloud?utm_source=github&utm_medium=referral&utm_campaign=20260420_v5_github&utm_content=resources_main
- Pricing: https://seedance2.cloud/pricing?utm_source=github&utm_medium=referral&utm_campaign=20260420_v5_github&utm_content=resources_pricing
- Gallery (examples): https://seedance2.cloud/showcases?utm_source=github&utm_medium=referral&utm_campaign=20260420_v5_github&utm_content=resources_showcases
- Blog: https://seedance2.cloud/blog?utm_source=github&utm_medium=referral&utm_campaign=20260420_v5_github&utm_content=resources_blog
- Contact: hello@seedance2.cloud
This README and the profile assets in this repository are released under MIT. The Seedance 2 Cloud platform itself is proprietary.
Seedance 2 Cloud is an independent implementation built on the publicly available Seedance 2.0 model. Not affiliated with ByteDance or its subsidiaries.