Create images with flux lora style model

Click the "style" icon in the sidebar. The 'FLUX-1-dev' model is recommended for detailed and high-quality images.

LoRA models are specialized styles trained to generate specific artistic outputs. Select a LoRA model from the 'Select Lora Models' section. You can also add trigger words automatically to ensure the LoRA model influences the final output. Each LoRA model has its own unique trigger words that will be applied to your prompt.

If you'd like to base your generated image on an existing image, use the 'Upload Image' option. Once uploaded, you can control the strength of image-to-image guidance using the 'Image Denoise' slider. A value of 1.0 completely remakes the image, while 0.0 keeps the original image almost unchanged. The default setting is 0.85, which balances transformation and preservation.

Type your image description in the 'Prompt' field. The more detailed and specific the prompt, the better the image quality will be. Use descriptive words to guide the style and content of the image. Feel free to use the magic prompt generator for suggestions if needed.

Choose an 'Aspect Ratio' that suits your image’s purpose. Different aspect ratios are suitable for various uses like social media posts (1:1), banners (16:9), or posters (3:4). Customize other advanced settings such as Seed, Guidance Scale, and Inference Steps. The Seed controls randomness, Guidance Scale fine-tunes how closely the image follows your prompt, and Inference Steps control the processing time and image quality.

Click the 'Generate' button to create your image. The cost will be deducted in credits. Once the image is generated, you can download it and review the results.