How to Generate Model Images with Style Reference
Step 1: Select the Style tab on the model generation section
Select the Style tab on the model generation section to start generating model images with a style reference.

Step 2: Select a similarity level
Select a similarity level. More similar = more similar to the reference image, vice versa. This similarity corresponds to the Similarity controls in the Advanced section. The higher the score, makes the image more closely match the text prompt. You can also customize the range yourself. Increasing the number of the output image will add more interval prompt strengths in the similarity control section.
Step 3: Select a LoRA model
Select a LoRA model. This is to indicate your output image should look like the model ID’s face you’ve trained. Toggling off the LoRA model selector will skip the LoRA selection and generate a random AI generated face with the same style as the reference image.

Step 4: Auto Generate Prompt
Click the generate prompt button to generate the prompt based on the reference images. If you select a similarity level as less similar, you can change the prompts to something more different from the reference image.

Step 5: Submit the request
Click the generate button to submit the request. The rest of the steps are the same as Attribute and Face Generation.
Pro Tip
Configure Model Attribute module only appear when selected Less Similar.
FAQs
What is Virtual Try-On?
HuHu AI’s virtual try-on tool revolutionizes how fashion companies showcase products using advanced AI to create realistic visuals of garments on customizable virtual models with unmatched accuracy and style.
Are the virtual try on photos licensed for commercial use?
Yes, you can use the generated virtual try on shots for commercial purposes. HuHu offers copyright-free preset AI models with different skin tones, ages, and features.
Please ensure that any images you upload (e.g., product photos) do not infringe third-party rights. We also recommend checking local laws before commercial use, as users are fully responsible for content compliance.
What are the advantages of AI virtual try-on compared to traditional photoshoots?
AI virtual try-on significantly lowers production costs, delivers results in minutes, and allows full control over model attributes like pose, body type, and styling.
How can I create my own AI models for virtual try on?
You can fully customize your AI fashion models using our model generator feature. Choose from a wide range of options including body shape, skin tone, hairstyle, facial features, and more. You can also upload a reference photo to generate an AI model with a similar look and style.
Related Tutorials
How to Generate Model Image with Specific Pose and Background
Learn how to create a reference model image by compositing background and foreground layers for use as a style reference in HuHu AI model generation.
How to Generate Model Images with Attributes
Learn how to generate AI model images using attribute controls in HuHu AI Studio, from selecting attributes to fine-tuning prompts and advanced settings.
How to Generate Model Images with Face Reference
Learn how to generate AI model images using face references in HuHu AI Studio, including uploading faces, adding background references, and customizing attributes.