diffusers/pr_14865/en/using-diffusers/controlnet.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;ControlNet&quot;,&quot;local&quot;:&quot;controlnet&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Multi-ControlNet&quot;,&quot;local&quot;:&quot;multi-controlnet&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Guess mode&quot;,&quot;local&quot;:&quot;guess-mode&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
		<link href="/docs/diffusers/main/en/_app/immutable/entry/start.Dp9eb7Q3.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/cl-n0R7Q.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/DK803DsY.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/entry/app.CVP8BBWW.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/DTwaC60R.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/BTASUwav.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/DnVIJfC5.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/0.AKqUuSqj.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/BzOvRKAw.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/361.BAUCLQAZ.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Bqo1LwON.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/BaXzLV62.js" rel="modulepreload">
		<!--18lijse--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;ControlNet&quot;,&quot;local&quot;:&quot;controlnet&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Multi-ControlNet&quot;,&quot;local&quot;:&quot;multi-controlnet&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Guess mode&quot;,&quot;local&quot;:&quot;guess-mode&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/><!---->
		<link href="/docs/diffusers/main/en/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----></span> <span>Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg><!----></button></div> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="controlnet" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#controlnet"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>ControlNet</span></h1><!--]--><!----> <p><a href="https://huggingface.co/papers/2302.05543" rel="nofollow">ControlNet</a> steers a pretrained diffusion model with a structural control image (edges, depth, or pose) while you keep a text prompt. It freezes the base model and adds a parallel network whose residuals guide the denoiser, so you get layout control without fine-tuning.</p> <p>Load a ControlNet for the control you need (for example canny), then pass it as <code>controlnet=</code> to <a href="/docs/diffusers/pr_14865/en/api/pipelines/overview#diffusers.DiffusionPipeline.from_pretrained">from_pretrained()</a>. Use <code>controlnet_conditioning_scale</code> to set how strongly the control steers generation.</p> <blockquote class="tip"><p>ControlNets are available to many models such as <a href="../api/pipelines/controlnet_flux">Flux</a>, <a href="../api/pipelines/controlnet_hunyuandit">Hunyuan-DiT</a>, <a href="../api/pipelines/controlnet_sd3">Stable Diffusion 3</a>, and more. The examples in this guide use Flux and Stable Diffusion XL. For T2I-Adapter on older Stable Diffusion–family checkpoints, see <a href="./legacy_adapters#t2i-adapter">Legacy adapters</a>. Prefer ControlNet for new controllable-generation work.</p></blockquote> <p>The tabs below show ControlNet for text-to-image, image-to-image, and inpainting.</p> <div class="flex space-x-2 items-center my-1.5 mr-8 h-7 !pl-0 -mx-3 md:mx-0"><!--[--><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd border-gray-800 bg-black dark:bg-gray-700 text-white">text-to-image</div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">image-to-image</div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">inpainting</div><!--]--></div> <div class="language-select"><!--[0--><p>Generate a canny image with <a href="https://github.com/opencv/opencv-python" rel="nofollow">opencv-python</a>.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> cv2
<span class="hljs-keyword">import</span> numpy <span class="hljs-keyword">as</span> np
<span class="hljs-keyword">from</span> PIL <span class="hljs-keyword">import</span> Image
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

original_image = load_image(
    <span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png&quot;</span>
)

image = np.array(original_image)

low_threshold = <span class="hljs-number">100</span>
high_threshold = <span class="hljs-number">200</span>

image = cv2.Canny(image, low_threshold, high_threshold)
image = image[:, :, <span class="hljs-literal">None</span>]
image = np.concatenate([image, image, image], axis=<span class="hljs-number">2</span>)
canny_image = Image.fromarray(image)<!----></pre></div><!----> <p>Pass the canny image to the pipeline. Use the <code>controlnet_conditioning_scale</code> parameter to determine how much weight to assign to the control.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> FluxControlNetPipeline, FluxControlNetModel

controlnet = FluxControlNetModel.from_pretrained(
    <span class="hljs-string">&quot;InstantX/FLUX.1-dev-Controlnet-Canny&quot;</span>, dtype=torch.bfloat16
)
pipeline = FluxControlNetPipeline.from_pretrained(
    <span class="hljs-string">&quot;black-forest-labs/FLUX.1-dev&quot;</span>, controlnet=controlnet, dtype=torch.bfloat16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

prompt = <span class="hljs-string">&quot;&quot;&quot;
A photorealistic overhead image of a cat reclining sideways in a flamingo pool floatie holding a margarita. 
The cat is floating leisurely in the pool and completely relaxed and happy.
&quot;&quot;&quot;</span>

pipeline(
    prompt, 
    control_image=canny_image,
    controlnet_conditioning_scale=<span class="hljs-number">0.5</span>,
    num_inference_steps=<span class="hljs-number">50</span>, 
    guidance_scale=<span class="hljs-number">3.5</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png" width="300" alt="Generated image (prompt only)"/> <figcaption style="text-align: center;">original image</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/canny-cat.png" width="300" alt="Control image (Canny edges)"/> <figcaption style="text-align: center;">canny image</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/canny-cat-generated.png" width="300" alt="Generated image (ControlNet + prompt)"/> <figcaption style="text-align: center;">generated image</figcaption></figure></div><!----><!--]--><!----> <!--[-1--><!--]--><!----> <!--[-1--><!--]--><!----><!----></div><!----> <!--[1--><h2 class="relative group"><a id="multi-controlnet" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#multi-controlnet"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Multi-ControlNet</span></h2><!--]--><!----> <p>Compose several ControlNets (for example canny + depth) by passing them as a list. Mask overlapping regions when you can, and tune each <code>controlnet_conditioning_scale</code>. Build <code>canny_image</code> with OpenCV Canny (as in text-to-image) and <code>depth_image</code> with DPT (as in image-to-image). Resize every control image to the pipeline size before you call the pipeline.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> cv2
<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">import</span> numpy <span class="hljs-keyword">as</span> np
<span class="hljs-keyword">from</span> PIL <span class="hljs-keyword">import</span> Image
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> DPTImageProcessor, DPTForDepthEstimation
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionXLControlNetPipeline, ControlNetModel, AutoencoderKL
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

original_image = load_image(
    <span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-enhanced-prompt.png&quot;</span>
)
image = np.array(original_image)
canny_image = Image.fromarray(
    np.concatenate([cv2.Canny(image, <span class="hljs-number">100</span>, <span class="hljs-number">200</span>)[:, :, <span class="hljs-literal">None</span>]] * <span class="hljs-number">3</span>, axis=<span class="hljs-number">2</span>)
)

device = <span class="hljs-string">&quot;cuda&quot;</span>  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
depth_estimator = DPTForDepthEstimation.from_pretrained(<span class="hljs-string">&quot;Intel/dpt-hybrid-midas&quot;</span>).to(device)
feature_extractor = DPTImageProcessor.from_pretrained(<span class="hljs-string">&quot;Intel/dpt-hybrid-midas&quot;</span>)

<span class="hljs-keyword">def</span> <span class="hljs-title function_">get_depth_map</span>(<span class="hljs-params">image</span>):
    image = feature_extractor(images=image, return_tensors=<span class="hljs-string">&quot;pt&quot;</span>).pixel_values.to(device)
    <span class="hljs-keyword">with</span> torch.no_grad(), torch.autocast(device):
        depth_map = depth_estimator(image).predicted_depth

    depth_map = torch.nn.functional.interpolate(
        depth_map.unsqueeze(<span class="hljs-number">1</span>),
        size=(<span class="hljs-number">1024</span>, <span class="hljs-number">1024</span>),
        mode=<span class="hljs-string">&quot;bicubic&quot;</span>,
        align_corners=<span class="hljs-literal">False</span>,
    )
    depth_min = torch.amin(depth_map, dim=[<span class="hljs-number">1</span>, <span class="hljs-number">2</span>, <span class="hljs-number">3</span>], keepdim=<span class="hljs-literal">True</span>)
    depth_max = torch.amax(depth_map, dim=[<span class="hljs-number">1</span>, <span class="hljs-number">2</span>, <span class="hljs-number">3</span>], keepdim=<span class="hljs-literal">True</span>)
    depth_map = (depth_map - depth_min) / (depth_max - depth_min)
    image = torch.cat([depth_map] * <span class="hljs-number">3</span>, dim=<span class="hljs-number">1</span>)
    image = image.permute(<span class="hljs-number">0</span>, <span class="hljs-number">2</span>, <span class="hljs-number">3</span>, <span class="hljs-number">1</span>).cpu().numpy()[<span class="hljs-number">0</span>]
    image = Image.fromarray((image * <span class="hljs-number">255.0</span>).clip(<span class="hljs-number">0</span>, <span class="hljs-number">255</span>).astype(np.uint8))
    <span class="hljs-keyword">return</span> image

depth_image = get_depth_map(original_image.resize((<span class="hljs-number">1024</span>, <span class="hljs-number">1024</span>)))

controlnets = [
    ControlNetModel.from_pretrained(
        <span class="hljs-string">&quot;diffusers/controlnet-canny-sdxl-1.0&quot;</span>, dtype=torch.float16
    ),
    ControlNetModel.from_pretrained(
        <span class="hljs-string">&quot;diffusers/controlnet-depth-sdxl-1.0-small&quot;</span>, dtype=torch.float16
    ),
]

vae = AutoencoderKL.from_pretrained(<span class="hljs-string">&quot;madebyollin/sdxl-vae-fp16-fix&quot;</span>, dtype=torch.float16)
pipeline = StableDiffusionXLControlNetPipeline.from_pretrained(
    <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>, controlnet=controlnets, vae=vae, dtype=torch.float16
).to(device)

prompt = <span class="hljs-string">&quot;&quot;&quot;
a relaxed rabbit sitting on a striped towel next to a pool with a tropical drink nearby, 
bright sunny day, vacation scene, 35mm photograph, film, professional, 4k, highly detailed
&quot;&quot;&quot;</span>
negative_prompt = <span class="hljs-string">&quot;lowres, bad anatomy, worst quality, low quality, deformed, ugly&quot;</span>

images = [canny_image.resize((<span class="hljs-number">1024</span>, <span class="hljs-number">1024</span>)), depth_image.resize((<span class="hljs-number">1024</span>, <span class="hljs-number">1024</span>))]

pipeline(
    prompt,
    negative_prompt=negative_prompt,
    image=images,
    num_inference_steps=<span class="hljs-number">100</span>,
    controlnet_conditioning_scale=[<span class="hljs-number">0.5</span>, <span class="hljs-number">0.5</span>],
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/canny-cat.png" width="300" alt="Canny image"/> <figcaption style="text-align: center;">canny image</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/multicontrolnet_depth.png" width="300" alt="Depth map"/> <figcaption style="text-align: center;">depth map</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/sdxl_multi_controlnet.png" width="300" alt="Generated image (ControlNet + prompt)"/> <figcaption style="text-align: center;">generated image</figcaption></figure></div> <!--[1--><h2 class="relative group"><a id="guess-mode" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#guess-mode"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span>Guess mode</span></h2><!--]--><!----> <p><a href="https://github.com/lllyasviel/ControlNet/discussions/188" rel="nofollow">Guess mode</a> generates an image from a control alone when you don’t have (or don’t want) a text prompt. It scales ControlNet residuals by block depth, from about <code>0.1</code> in early down blocks up to <code>1.0</code> at the mid block. That keeps early layers from over-constraining while deeper blocks still carry the structure.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copied</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionXLControlNetPipeline, ControlNetModel

controlnet = ControlNetModel.from_pretrained(
  <span class="hljs-string">&quot;diffusers/controlnet-canny-sdxl-1.0&quot;</span>, dtype=torch.float16
)
pipeline = StableDiffusionXLControlNetPipeline.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  controlnet=controlnet,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

canny_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/canny-cat.png&quot;</span>)
pipeline(
  <span class="hljs-string">&quot;&quot;</span>,
  image=canny_image,
  guess_mode=<span class="hljs-literal">True</span>
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/canny-cat.png" width="300" alt="Canny image"/> <figcaption style="text-align: center;">canny image</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guess_mode.png" width="300" alt="Generated image (Guess mode)"/> <figcaption style="text-align: center;">generated image</figcaption></figure></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/diffusers/blob/main/docs/source/en/using-diffusers/controlnet.md" target="_blank"><svg class="mr-1" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg><!----> <span><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]-->
			
			<script>
				{
					__sveltekit_1a0mh15 = {
						base: "/docs/diffusers/main/en",
						assets: "/docs/diffusers/main/en"
					};

					const element = document.currentScript.parentElement;

					Promise.all([
						import("/docs/diffusers/main/en/_app/immutable/entry/start.Dp9eb7Q3.js"),
						import("/docs/diffusers/main/en/_app/immutable/entry/app.CVP8BBWW.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 361],
							data: [null,null],
							form: null,
							error: null
						});
					});
				}
			</script>