SDXL
SDXL (Stable Diffusion XL) is a large-scale text-to-image generation model released by Stability AI and an upgraded version of the Stable Diffusion series. It employs a dual text-encoder architecture with a substantially larger parameter count, enabling the generation of higher-resolution images with richer detail. The model also demonstrates stronger semantic understanding of complex prompts and is widely used in AI art creation and design assistance applications.
Timeline (last 90 days)
The U-Net portion of SD 1.5 has about 860 million parameters, and SDXL has about 2.6 billion parameters.
SDXL (RealVisXL V5.0) plus ControlNet canny/depth can precisely preserve pose frame by frame and produce photorealistic results within seconds, but it is equally faithful to input errors (gray placeholder equipment and hands not gripping the bar will be copied), and prompts longer than 77 tokens are silently truncated.
In the autonomous art direction agent built by the developer, the planner generates shot descriptions, SDXL (Juggernaut XL v9) renders images, a vision model evaluates the output and writes correction instructions, and the loop reruns until the critic model passes it.
Stable Diffusion 1.5 has about 0.9B parameters, SDXL about 2.6B, and Flux.1-dev 12B.
By default, after generating one image, the SDXL checkpoint, CLIP text encoder, and VAE remain resident in VRAM, occupying about 8GB; after calling /free, VRAM falls back to a baseline of about 2.2GB.
In ComfyUI, running SDXL alone to generate one image takes about 15 to 30 seconds.
SDXL generating a 1024×1024 image occupies approximately 6–8 GB of VRAM.
An Ollama 14b model combined with SDXL 1024×1024 generation requires approximately 14.5 GB of VRAM, which is tight but barely feasible on a 16 GB card.
The SDXL base model alone occupies approximately 6–7 GB of VRAM, and peak VRAM can easily exceed 20 GB when simultaneously loading the Refiner, multiple ControlNet weights (approximately 1.4 GB each), and a high-precision VAE.
SD XL introduced a dual-encoder architecture (CLIP ViT-L + OpenCLIP ViT-bigG), improving semantic capture for long prompts.
23 more timeline events
All Facts (20)
SDXL采用了双CLIP编码器（OpenCLIP ViT-bigG和OpenAI CLIP ViT-L）的设计
90%Unverifiedsd-scripts支持LoRA、LyCORIS（LoHa、LoKr等）、Textual Inversion、DreamBooth等微调方法，并支持SD 1.5、SDXL和SD3
90%UnverifiedSD1.5使用768维条件向量，SDXL使用1024维条件向量
85%UnverifiedSD 1.5系列模型约需4-6G显存，SDXL系列通常需要8-12G显存
85%UnverifiedThe SDXL model requires approximately 6–8 GB of VRAM at FP16 precision, and adding auxiliary modules such as ControlNet or LoRA pushes that beyond 10 GB.
60%UnverifiedBetween 2023 and 2024, the image generation field experienced an architectural leap from Stable Diffusion 1.5 to SDXL and then to SD3.
60%UnverifiedSDXL（Stable Diffusion XL）是Stability AI推出的高分辨率图像生成模型，其完整管线包含Base模型和Refiner模型两个阶段
60%UnverifiedThe U-Net portion of SD 1.5 has about 860 million parameters, and SDXL has about 2.6 billion parameters.
50%UnverifiedSDXL (RealVisXL V5.0) plus ControlNet canny/depth can precisely preserve pose frame by frame and produce photorealistic results within seconds, but it is equally faithful to input errors (gray placeholder equipment and hands not gripping the bar will be copied), and prompts longer than 77 tokens are silently truncated.
50%UnverifiedIn the autonomous art direction agent built by the developer, the planner generates shot descriptions, SDXL (Juggernaut XL v9) renders images, a vision model evaluates the output and writes correction instructions, and the loop reruns until the critic model passes it.
50%UnverifiedStable Diffusion 1.5 has about 0.9B parameters, SDXL about 2.6B, and Flux.1-dev 12B.
50%UnverifiedIn ComfyUI, running SDXL alone to generate one image takes about 15 to 30 seconds.
50%UnverifiedBy default, after generating one image, the SDXL checkpoint, CLIP text encoder, and VAE remain resident in VRAM, occupying about 8GB; after calling /free, VRAM falls back to a baseline of about 2.2GB.
50%UnverifiedAn Ollama 14b model combined with SDXL 1024×1024 generation requires approximately 14.5 GB of VRAM, which is tight but barely feasible on a 16 GB card.
50%UnverifiedSDXL generating a 1024×1024 image occupies approximately 6–8 GB of VRAM.
50%UnverifiedThe SDXL base model alone occupies approximately 6–7 GB of VRAM, and peak VRAM can easily exceed 20 GB when simultaneously loading the Refiner, multiple ControlNet weights (approximately 1.4 GB each), and a high-precision VAE.
50%UnverifiedSD XL introduced a dual-encoder architecture (CLIP ViT-L + OpenCLIP ViT-bigG), improving semantic capture for long prompts.
50%UnverifiedAccording to the publisher's description, Nanosaur 2 outperforms the SDXL model in many comparisons, despite having far fewer parameters than SDXL.
50%UnverifiedThe score gaps between subsequent models such as Flux and SDXL on public benchmarks have already become quite limited.
50%UnverifiedRunning mainstream image generation models (such as SDXL) requires at least 8GB of VRAM
50%