Skip to content
Sign in
  • Browse all toolsFilter by category and pricing
  • Browse categoriesWriting, code, image, audio and more
  • Compare toolsPut up to three side by side
  • Read the blogGuides and comparisons

ControlNet Pose

Generate images of humans using Stable Diffusion, a text prompt and poses detected from an input image.

ControlNet Pose is a pose-guided image generation model that combines Stable Diffusion, a text prompt and human poses detected from an input image. ControlNet Pose serves people generating images of humans and is available through paid Replicate runs or as an open-source model runnable locally with Docker.

Updated 15 Nov 20266 sections

At a glance

Pricing
PaidControlNet Pose costs approximately $0.15 per run on Replicate, varying by input. The open-source model can also run on a user's own computer with Docker.
Added
15 Nov 2026
ControlNet Pose website screenshot

What is ControlNet Pose?

ControlNet Pose is an open-source model for generating images using human poses from a reference image and a text prompt. ControlNet Pose adapts Stable Diffusion with ControlNet, adding a human pose map as an input condition. The model is intended for image generation involving humans, where the pose in an existing image provides guidance beyond the written prompt.

What ControlNet Pose can do

ControlNet Pose combines image-based pose information with text-based generation instructions. The source image supplies the human pose, while the prompt describes the image to generate in the same manner as a Stable Diffusion prompt. Openpose handles pose detection automatically, so the documented workflow begins with an image rather than requiring a separately prepared pose map.

  • ControlNet Pose detects human poses in an input image using Openpose.
  • ControlNet Pose uses a human pose map to condition Stable Diffusion.
  • ControlNet Pose accepts a text prompt alongside the input image.
  • ControlNet Pose generates an output image from the combined inputs.

How ControlNet Pose works

ControlNet Pose follows the documented workflow of supplying an image and prompting the model as for Stable Diffusion. Openpose detects the human pose, and ControlNet adds that pose information to the generation process. The Replicate page provides Playground, API and examples, while a GitHub link points to the original model and code. Local execution is also available through Docker.

Who ControlNet Pose is for

ControlNet Pose is relevant to people generating images of humans who want an input image to guide the pose. The model addresses a specific form of conditional image generation: combining a detected human pose with a written description. The source presents other ControlNet models for drawings, edges, depth and other conditions separately, rather than describing those methods as capabilities of ControlNet Pose.

  • ControlNet Pose supports generating human images with pose guidance from a reference image.
  • ControlNet Pose supports exploring text prompts alongside an image-derived human pose.
  • ControlNet Pose supports local experimentation with the open-source model through Docker.

ControlNet Pose pricing and plans

ControlNet Pose uses per-run pricing on Replicate rather than a subscription plan described on the page. The stated approximate cost is $0.15 per run, or 6 runs per $1, with costs varying according to the inputs. The source also identifies the model as open source and says users can run the model on their own computer with Docker.

ControlNet Pose strengths and limitations

ControlNet Pose provides a concrete additional control for Stable Diffusion: a human pose detected from an image. Automatic pose detection and the choice between hosted execution and local Docker execution are practical strengths. On Replicate, the model runs on Nvidia A100 (80GB) GPU hardware, and predictions typically complete within 108 seconds. The page cautions that prediction time varies significantly depending on the inputs.

  • ControlNet Pose hosted runs incur a cost that varies with the inputs.
  • ControlNet Pose prediction times are not fixed.
  • ControlNet Pose requires Docker for the stated local execution option.

ControlNet Pose FAQ

What is ControlNet Pose?

ControlNet Pose is an open-source image generation model that adapts Stable Diffusion to use a human pose map from an input image alongside a text prompt. Openpose detects the pose automatically.

Is ControlNet Pose free?

ControlNet Pose costs approximately $0.15 per run on Replicate, with the actual cost depending on the inputs. ControlNet Pose is also open source and can run on a user's own computer with Docker.

How do you use ControlNet Pose?

ControlNet Pose takes an input image and a text prompt written as for Stable Diffusion. Openpose detects the human pose, and the model uses the resulting pose map together with the prompt to generate an output image.

Can ControlNet Pose run locally?

ControlNet Pose can run on a user's own computer with Docker, according to the model page. The page also links to the original model and code on GitHub and offers hosted execution through Replicate.

How long does ControlNet Pose take to generate an image?

ControlNet Pose predictions typically complete within 108 seconds on Replicate, where the model runs on Nvidia A100 (80GB) GPU hardware. The page states that prediction time varies significantly based on the inputs.

Does ControlNet Pose have an API?

ControlNet Pose has an API option on its Replicate model page, alongside Playground and examples. The documented model workflow uses an input image and a text prompt, with Openpose detecting the human pose automatically.

From the community

No reviews yet

Be the first to review this tool. What you write helps the next person choose.

More tools like ControlNet Pose

Text-To-Pokemon

www.replicate.com

Generate Pokémon characters from text prompts with a fine-tuned Stable Diffusion model.

Paid

QR Diffusion

www.qrdiffusion.com

Generate artistic QR codes with AI, editable templates, dynamic links and performance analytics.

Freemium

ArtBot

www.tinybots.net

Create and edit AI images with Stable Diffusion through the free AI Horde network.

Free

GRAVITI Diffus

diffus.graviti.com

Browser-based AI image generation with custom models, ControlNet, inpainting, and face enhancement.

Paid