10 actions in 10 pieces of software · ranked by words, meaning and measured trust · 650 ms

1. 1cjwbw/kandinsky-2-2-controlnet-depthverified and signedpassport

   modelrecord

   • cjwbw/kandinsky-2-2-controlnet-depth

     Kandinsky Image Generation with ControlNet Conditioning

2. 2stable-diffusion-3.5-large-controlnet-depthverified and signedpassport

   modelrecord10,866 downloads/month

   • text-to-image

     Task: text-to-image

3. 3cjwbw/depth-anythingverified and signedpassport

   modelrecord

   • cjwbw/depth-anything

     Highly practical solution for robust monocular depth estimation by training on a combination of 1.5M labeled images and 62M+ unlabeled images

4. 4lucataco/sdxl-controlnet-depthverified and signedpassport

   modelrecord

   • lucataco/sdxl-controlnet-depth

     SDXL ControlNet - Depth

5. 5cjwbw/zoedepthverified and signedpassport

   modelrecord

   • cjwbw/zoedepth

     ZoeDepth: Combining relative and metric depth

6. 6cjwbw/midasverified and signedpassport

   modelrecord

   • cjwbw/midas

     Robust Monocular Depth Estimation

7. 7jagilley/controlnet-depth2imgverified and signedpassport

   modelrecord

   • jagilley/controlnet-depth2img

     Modify images using depth maps

8. 8ai-forever/kandinsky-2.2verified and signedpassport

   modelrecord

   • ai-forever/kandinsky-2.2

     multilingual text2image latent diffusion model

9. 9xlabs-ai/flux-dev-controlnetverified and signedpassport

   modelrecord

   • xlabs-ai/flux-dev-controlnet

     XLabs v3 canny, depth and soft edge controlnets for Flux.1 Dev

10. 10Krea-2 Depth ControlNetverified and signedpassport

    MCP serverrecord

    • krea2_psx_scrambler_generate

      Generate a new image that keeps the 3D structure of an input image. Extracts a depth map from the input image with Depth-Anything-V2 and generates a new image following the same depth/composition but with the content and style described by the prompt (Krea-2-Turbo, 8-step). If a user passes a file a