FAQ

We offer two models. Core is a streamlined model ideal for quick generations at 3 credits per image — just enter a prompt, pick a style, and choose an aspect ratio. SD3 gives you access to the Stable Diffusion 3.5 family: Large (6 credits), Large Turbo (4 credits), Medium (3 credits), and Flash (2 credits). SD3 also supports image-to-image generation, CFG scale control, and more advanced settings.

Be specific about the subject, composition, lighting, and mood you want. For example, instead of "a cat", try "a tabby cat sitting on a windowsill at golden hour, soft bokeh background". The more detail you provide, the closer the result will match your vision. You can also combine a written prompt with an art style preset for even better results.

Both models let you choose from a curated set of style presets — including 3D Model, Analog Film, Anime, Cinematic, Comic Book, Digital Art, Fantasy Art, Isometric, Line Art, Low Poly, Neon Punk, Origami, Photographic, and Pixel Art. Selecting a preset guides the AI toward that aesthetic. If you leave it unset, the AI relies entirely on your text prompt.

CFG (Classifier-Free Guidance) scale controls how strictly the AI follows your prompt. The Large and Medium models default to 4, while Turbo and Flash default to 1. Higher values produce results closer to your description but may look less natural. Lower values give the AI more creative freedom. Adjust this when you want to fine-tune the balance between prompt accuracy and artistic variation.

Choose from nine aspect ratios to match your intended use: 1:1 (square), 16:9 and 9:16 (widescreen / vertical), 21:9 and 9:21 (ultrawide / ultra-tall), 3:2 and 2:3, or 5:4 and 4:5. Picking the right ratio upfront avoids the need to crop later and ensures your image is optimized for its destination — social media, print, or web.

SD3 lets you upload a reference image and use it as a starting point. The strength slider controls how much the output diverges from the original — a low strength keeps it close, while a high strength reimagines it more freely. This is great for style transfers, variations, or iterating on an existing concept.

Each generation costs a set number of credits depending on the model: Core costs 3 credits, and SD3 ranges from 2 credits (Flash) to 6 credits (Large). Visit the Pricing page for details on credit bundles and plans.

AI generation is inherently non-deterministic. Even with the same prompt and settings, outputs can differ between runs. If a result isn't quite right, try rephrasing your prompt, switching style presets, or adjusting the CFG scale. Experimentation is part of the creative process.

You may use generated images for commercial purposes. However, AI regulations vary by jurisdiction and are evolving rapidly — make sure to stay informed about the rules that apply in your country.