# LumiYing

> AI video and image generation platform offering top-tier video and image models — Seedance 2.0, Kling 3.0, GPT Image 2, Midjourney, and more — from a single workspace with credit-based pricing.

## Docs

- [LumiYing Homepage](https://lumiying.com): AI video and image generation platform. Supports text, image, video, and audio inputs across multiple models. Features multimodal engine, audio-visual sync, and dynamic action generation.
- [All AI Models](https://lumiying.com/models): Complete catalog of video and image generation models available on LumiYing, including Seedance 2.0, Kling 3.0, GPT Image 2, Midjourney, Seedream, and Nano Banana series.
- [What is Seedance 2.0](https://lumiying.com/resource/what-is-seedance-2): Overview of ByteDance's multimodal AI video model. Covers 4-modality input (image, video, audio, text), native audio generation, specifications, use cases, and feature comparison with competitors.
- [Seedance 2.0 Tutorial](https://lumiying.com/resource/seedance-2-tutorial): Step-by-step guide covering the @tag reference system, prompt structure (Subject + Action + Scene + Camera + Style), settings configuration, common mistakes, and credit-saving workflow tips.
- [Seedance 2.0 API Guide](https://lumiying.com/resource/seedance-2-api): API preview documentation for programmatic video generation. Covers multimodal input specs, file format requirements, generation limits, and planned REST endpoint capabilities.
- [Seedance 2.0 Pricing](https://lumiying.com/resource/seedance-2-pricing): Expected pricing structure including tiered plans (Free, Starter, Pro, Enterprise), pay-as-you-go credits, API pricing, and volume discounts.

## Video Models

- [Seedance 2.0](https://lumiying.com/models/seedance-2-0): ByteDance's flagship 4-modality video generator. Accepts up to 12 files (9 images, 3 videos, 3 audio clips) per request. 720p at 24fps, 4–15 second duration, native stereo audio, 6 aspect ratios.
- [Seedance 2.0 Fast](https://lumiying.com/models/seedance-2-0-fast): Speed-optimized variant of Seedance 2.0. 2x faster generation with lower credit cost: 10 credits/sec without video input, or 6 credits/sec with video input. Same 4-modality input and duration options. Ideal for rapid iteration and prototyping.
- [Kling 3.0](https://lumiying.com/models/kling-3-0): Kuaishou's director-style AI video model for 3-15 second clips from text, start/end frames, or motion-control references. Supports 720p/1080p output, multiple aspect ratios, and optional native sound.
- [Kling 3.0 Omni](https://lumiying.com/models/kling-3-0-omni): Multi-reference Kling video model for reference videos, video editing, first/last frame generation, 720p/1080p output, and optional sound for non-video-reference workflows.
- [Seedance 1.5 Pro](https://lumiying.com/models/seedance-1-5-pro): Multi-shot AI video with cinematic camera control and 8-language lip-sync support.

## Image Models

- [GPT Image 2](https://lumiying.com/models/gpt-image-2): GPT-powered AI image generator for text-to-image creation and reference-image editing. Optimized for prompt adherence, text rendering, realistic detail, and 3 fixed aspect ratios.
- [Nano Banana 2](https://lumiying.com/models/nano-banana-2): Google's fastest AI image generator. Pro quality at flash speed, up to 4K resolution output.
- [Nano Banana Pro](https://lumiying.com/models/nano-banana-pro): Google's highest-quality image model. 4K output, brand control features, supports up to 14 reference images.
- [Midjourney](https://lumiying.com/models/midjourney): Industry standard for AI art. V7 with stylize control and text rendering capabilities.
- [Seedream 5.0 Lite](https://lumiying.com/models/seedream-5-0-lite): ByteDance's thinking image model with visual reasoning, RAG, and style transfer capabilities.
- [Seedream 4.5](https://lumiying.com/models/seedream-4-5): ByteDance's 4K flagship image model. Best-in-class text rendering with 9.6/10 face consistency score.

## Resources

- [How to Bypass Seedance 2.0 Face Detection](https://lumiying.com/resource/bypass-seedance-2-face-detection): Workaround guide for face detection rejections. Covers 4 methods (grid overlay, cropping, downscaling, style transfer) with step-by-step instructions and troubleshooting.
- [AI Video Showcase](https://lumiying.com/showcase): Gallery of AI-generated videos created on LumiYing. Cinematic scenes, dynamic action, character animation, and creative storytelling examples with prompts.
- [Pricing](https://lumiying.com/pricing): Credit package page for LumiYing video and image generation.

## Tools

- [Image Grid Overlay Tool](https://lumiying.com/tools/image-grid-overlay): Free browser-based tool to add grid overlays to images. Used to bypass Seedance 2.0 face detection. Supports PNG, JPG, WebP. No signup required, fully client-side processing.
- [Scenery Overlay Tool](https://lumiying.com/tools/scenery-overlay): Free browser-based tool to blend scenery or texture overlays onto portrait references before Seedance 2.0 generation.

## Key Facts

- LumiYing is an AI video and image generation platform
- Offers 9 video models and 6 image models from multiple providers (ByteDance, Google, Kuaishou, Midjourney)
- Credit-based pricing system — credits never expire
- Supports 4-modality input: text, image, video, and audio
- Video output ranges from 720p to 4K depending on model
- Video duration ranges from 4 to 15 seconds per generation depending on model
- Multiple models support native audio generation synchronized with video
- All models available to every user — no model locked behind specific tier
- The /models catalog is the canonical list of currently supported LumiYing generation models

## Contact

- Website: https://lumiying.com
- App: https://lumiying.com/ai-tool/home