<h2 id="stable-diffusion-3.5-controlnets-ol2vzj840">Stable Diffusion 3.5 ControlNets</h2>Originally <a rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-3.5-controlnets">posted on Huggingface</a>.The following control types are available:<ul><li>Canny - Use a Canny edge map to guide the structure of the generated image. This is especially useful for illustrations, but works with all styles.</li><li>Depth - use a depth map, generated by DepthFM, to guide generation. Some example use cases include generating architectural renderings, or texturing 3D assets.</li><li>Blur - can be used to perform extremely high fidelity upscaling. A common use case is to tile an input image, apply the ControlNet to each tile, and merge the tiles to produce a higher resolution image. A more in-depth description of this use case is <a target="_blank" rel="ugc" href="https://github.com/lllyasviel/ControlNet-v1-1-nightly/issues/125">here</a>.<ul><li>For Comfy users, <a target="_blank" rel="ugc" href="https://github.com/shiimizu/ComfyUI-TiledDiffusion">this extension</a> provides support.</li><li>We recommend tiling the image at a tile size between 128 and 512.</li></ul></li></ul>All currently released ControlNets are compatible only with Stable Diffusion 3.5 Large (8b).

Files with SHA256: 4bc5cf949f6501a4bd125c6c1190e8fba0f1471f5ce36e9ebd1114867abedd4c