--- license: creativeml-openrail-m tags: - pytorch - diffusers - stable-diffusion - text-to-image - diffusion-models-class - dreambooth-hackathon - wildcard widget: - text: illustration of a haojing people playing computer --- # DreamBooth model for the haojing concept trained by cccccccccccccccccc. This is a Stable Diffusion model fine-tuned on the haojing concept with DreamBooth. It can be used by modifying the `instance_prompt`: **a photo of haojing people** This model was created as part of the DreamBooth Hackathon 🔥. Visit the [organisation page](https://huggingface.co/dreambooth-hackathon) for instructions on how to take part! ## Description This is a Stable Diffusion model fine-tuned on `people` images for the wildcard theme, for the Hugging Face DreamBooth Hackathon, from the HF CN Community, corporated with the HeyWhale. ## Usage ```python from diffusers import StableDiffusionPipeline pipeline = StableDiffusionPipeline.from_pretrained('cccccccccccccccccc/haojing-people-heywhale') image = pipeline().images[0] image ```