The simplest way to self-host Dreamshaper XL. Launch a dedicated cloud GPU server running Lab Station OS to download and serve the model using any compatible app or framework.
Direct Download
Download model weights for local inference. Must be used with a compatible app, notebook, or codebase. May run slowly, or not work at all, depending on your system resources, particularly GPU(s) and available VRAM.
Forge is a platform built on top of Stable Diffusion WebUI to make development easier, optimize resource management, speed up inference, and study experimental features.
Train your own LoRAs and finetunes for Stable Diffusion and Flux using this popular GUI for the Kohya trainers.
Model Report
Lykon / Dreamshaper XL
DreamShaper XL is a Stable Diffusion XL-based model with Lightning and Turbo variants optimized for different sampling methods. It excels at artistic, anime, and manga generation, featuring enhanced edge definition and improved human subject rendering. Supports tiled upscaling for higher resolution outputs.
Explore the Future of AI
Your server, your data, under your control
DreamShaper XL is a versatile Stable Diffusion XL (SDXL) model designed for generating high-quality images across multiple artistic styles, from photorealistic to anime and manga. The model demonstrates particular strength in generating detailed human figures, sharp edges, and specific subjects like dragons.
Model Variants and Technical Specifications
DreamShaper XL comes in several variants, each optimized for different use cases:
Lightning Variant: Optimized for rapid generation using 3-6 sampling steps at a CFG scale of 2
Turbo Variant: Designed for slightly longer generation times using 4-8 sampling steps at a CFG scale of 2
SFW Variants: Safe For Work versions of both Lightning and Turbo
Legacy Versions: Earlier alpha versions based on SDXL 0.9 and 1.0
Both Lightning and Turbo variants utilize advanced sampling algorithms - Lightning DPM++ SDE and Turbo DPM++ SDE respectively. For optimal results, both variants should use the DPM++ SDE Karras sampler. The model supports high-resolution fixes and tiled upscaling capabilities, making it suitable for generating large-scale, detailed images.
Features and Capabilities
DreamShaper XL excels in multiple areas:
Photorealistic image generation
Artistic and stylized renderings
Anime and manga-style artwork
Detailed human figure generation
Sharp edge definition
Specialized subject matter (e.g., dragons)
The model shows significant improvements over previous versions, particularly in image quality and specific subject rendering. Users can find detailed comparisons with other SDXL models in the model comparison post.
Licensing and Usage
DreamShaper XL is released under the CreativeML Open RAIL++-M license. However, the Turbo variant has specific commercial restrictions that require permission from StabilityAI for commercial use. The base model builds upon SDXL 1.0, whose license terms can be found in the SDXL 1.0 License.
