A model's usual training resolution, used as a starting point for image size.
Stable Diffusion 1.5 commonly uses 512×512. SDXL commonly uses about 1024×1024 and was also trained with other aspect ratios. Individual models can have different recommendations.Start with the selected model's size presets or published suggestions. Very small dimensions may lose important features; very large dimensions can introduce repeated subjects or distorted anatomy.The examples show one request at several sizes. Changes in framing or missing subjects describe those results, rather than a fixed threshold for every image.The studio can update an untouched size when you change model families. Check the dimensions before generating. For larger output, see .
The examples on this page were made from one prompt, one model and one seed, changing only the setting shown. Other pictures, models and settings respond differently; compare on your own request with the seed and the other settings fixed.