Getting started guide on how to use different models

Click the "creation" icon in the sidebar. From the 'Model' dropdown, choose the AI model that best suits your needs. For fast results, select the 'FLUX-schnell' model, which offers a free trial for quick image generation. For high-quality images, the 'FLUX-dev' model is recommended, though it may take longer.

Enter your prompt in the 'Prompt' field. The more detailed and specific your prompt, the better the quality and relevance of the generated image. Use descriptive words to enhance the results, and feel free to use the magic prompt generator for assistance.

Customize the generation by selecting the 'Aspect Ratio'. Choose the appropriate ratio based on your image’s purpose—1:1 for social media posts, 16:9 for banners, and 3:4 for posters. You can also set a seed (or randomize it), select the 'Output Format' (PNG is recommended), and adjust the 'Number of Inference Steps' to control image quality and processing time.

Click the 'Generate' button to create your image. If you have a subscription, you can use 'Pay Credits' for faster image generation and higher-quality outputs. The generated images can be used for commercial purposes once you use pay credits. After generation, you can download the image and review the results.

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.

Click the "character" icon in the sidebar. Begin by uploading a portrait image using the 'Upload Image' button. This feature requires an image as the foundation for generating consistent likeness across all output images.

In the 'Prompt' field, enter a detailed description of the character or subject in the image. Including specific features like hairstyle, clothing, or facial expression ensures that the generated images stay consistent with the original subject.

Choose the desired 'Aspect Ratio' for your image. This determines the dimensions of your output, with options such as 1:1 for square images or 16:9 for banners. Selecting the right aspect ratio depends on your use case for the image.

Use the 'Image Guide' slider to adjust how closely the generated image should follow the original. A lower value (closer to 0) retains more resemblance to the original image, while a higher value (closer to 4) allows more creative variations but might reduce likeness.

You can further customize your image generation by tweaking advanced settings like Seed, Guidance Scale, and Inference Steps. These control the randomization, prompt adherence, and overall processing time for your image.

Once your settings are finalized, click 'Generate'. This feature requires credits, and the amount will be deducted upon generation. After the image is created, you can download it and review the results.

Use the 'Upload Image' button to upload a reference image for the image-to-image process. This step is required to proceed with image generation. Adjust the 'Image Denoise' slider to control how much of the original image is preserved—a lower value retains more details, while a value closer to 1.0 remakes the image entirely.

Toggle the 'Enable ControlNet' option to use ControlNet. This feature enhances the preservation of elements from the original image. You can click the button next to the ControlNet option to view examples of different ControlNet effects and how they influence the final image.

Select the 'ControlNet Type' (e.g., Depth, Canny) and the corresponding 'Depth Preprocessor' (such as DepthAnything). These settings dictate how the structure and details from the uploaded image are retained during generation. Adjust the 'Control Strength' slider to control how strongly ControlNet influences the output. Different ControlNet settings are better suited to various styles—depth works well with values between 0.5 and 0.75.

Type a detailed description of the image you want to generate in the 'Prompt' field. Including specifics about the subject or style helps guide the AI in generating the image as close to your vision as possible.

If you want to apply a specific artistic style, select a LoRA model from the 'Select Lora Models' section. Only one LoRA model can be used when ControlNet is enabled. Adding the LoRA trigger word to your prompt will help apply the model’s style to the generated image.

Once all settings are in place, click the 'Generate' button. The number of credits required will be displayed. After the generation is complete, download your image and review the results.