I trained and released a Pose Control LoRA for Krea-2
retained text
Hey everyone! I built a Pose Control LoRA for Krea-2 and have fully released the LoRA checkpoints on Hugging Face. I also put together a ComfyUI workflow. It’s not fully optimized yet, but it’s usable for testing the model. Apart from that, I wrote a blog covering the decisions, data preparation, training process, evaluations, and some of the things I learned while building it. Hugging Face: [https://huggingface.co/adhit-420/Krea-2-PoseControl-LoRA-checkpoints](https://huggingface.co/adhit-420/Krea-2-PoseControl-LoRA-checkpoints) Blog: [https://s-adhit.github.io/blog/training-a-pose-control-lora-for-krea-2/](https://s-adhit.github.io/blog/training-a-pose-control-lora-for-krea-2/) Would love to hear your thoughts, feedback, or any interesting results if you try it! https://preview.redd.it/t2qq9kt1zfzh1.png?width=1536&format=png&auto=webp&s=44e4139cf65db3139e5638bfca584846c1013b65
document fetched 2026-11-04 00:30 · canonical https://reddit.com/r/StableDiffusion/comments/1xn145z/i_trained_and_released_a_pose_control_lora_for