thibaud

These are controlnet weights trained on stabilityai/stable-diffusion-xl-base-1.0 with OpenPose (v2) conditioning. You can find some example images in the following. (Image is from ComfyUI, you can drag and drop in Comfy to use it as workflow) Training data This checkpoint was first trained for 15,000 steps on laion 6a resized to a max minimum dimension of 768. Compute one 1xA100 machine (Thanks a lot HF🤗 to provide the compute!) Batch size Data parallel with a single gpu batch size of 2 with gradient accumulation 8.

Want to support my work: you can bought my Artbook: https://thibaud.art Here's the first version of controlnet for stablediffusion 2.1 Trained on a subset of laion/laion-art To use with Automatic1111: Download the ckpt files or safetensors ones Put it in extensions/sd-webui-controlnet/models in settings/controlnet, change cldmv15.yaml by cldmv21.yaml Enjoy To use ZoeDepth: You can use it with annotator depth/leres but it works better with ZoeDepth Annotator. My PR is not accepted yet but you can use my fork. My fork: https://github.com/thibaudart/sd-webui-controlnet The PR: https://github.com/Mikubill/sd-webui-controlnet/pull/655#issuecomment-1481724024 The model should not be used to intentionally create or disseminate images that create hostile or alienating environments for people. This includes generating images that people would foreseeably find disturbing, distressing, or offensive; or content that propagates historical or current stereotypes. Thanks https://huggingface.co/lllyasviel/ for the implementation and the release of 1.5 models. Thanks https://huggingface.co/p1atdev/ for the conversion script from ckpt to safetensors pruned & fp16 Models can't be sell, merge, distributed without prior writing agreement.