stable-diffusion.cpp System Requirements (CPU, RAM, Disk)

stable-diffusion.cpp is a lightweight, dependency-free inference engine for diffusion models such as Stable Diffusion, FLUX, Wan, Qwen Image and Z-Image, built in pure C/C++ on top of the ggml tensor library. It ships as a single native binary with no Python or PyTorch runtime, supports CPU, CUDA, Vulkan, Metal, SYCL and MUSA backends, and offers quantized GGUF model support, SD 1.5/2.x/SDXL/SD3.5, img2img, ControlNet, LCM and LoRA. Ideal for edge devices, servers and containers where a minimal footprint and fast startup matter.