Unverified50% confidenceFactExact time
The U-Net portion of SD 1.5 has about 860 million parameters, and SDXL has about 2.6 billion parameters.
1
Sources
50%
Confidence
Long-term
Relevance
10/29/2026
First Seen
Sources
Run 8K Images on 6GB VRAM! Hands-On Test of Fudan's Open-Source MOSS-Image Model
bilibiliAI绘画教学视频10/29/2026
Related Entities
Related Claims
UnverifiedSDXL increases the U-Net parameter count to 3.5 billion, adopts dual text encoders (OpenCLIP ViT-bigG and CLIP ViT-L), and uses a two-stage generation architecture (base + refiner).70% similarUnverifiedSDXL's parameter count grew from approximately 0.9B in SD1.5 to 6.6B.70% similarUnverifiedThe 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.70% similarUnverifiedSDXL 基础模型仅加载就需要 6-8GB 显存64% similarUnverified对于数十亿参数的SD模型，LoRA文件通常只有10-200MB64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1112060API
curl https://kongchang.com/api/v1/knowledge/claims/1112060MCP
get_claim(id=1112060)