01
Generates raster (PNG/JPG) images from text prompts
02
Generates true vector SVG files from text prompts. Competitors like Midjourney and OpenAI models do not offer native SVG output
03
First image model to support accurate long-text generation of any size and length within images, outperforming all alternatives
04
Supports resolutions up to 2048x1024 and 1024x2048, plus 1024x1024 and 10 intermediate sizes
05
Covers 15 aspect ratios: 1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16, 1:2, 2:1, 7:5, 5:7, 4:5, 5:4, 3:5, 5:3
06
Realistic image family: standard, B&W, hard flash, HDR, natural light, studio portrait, enterprise, motion blur
07
Digital illustration family: standard, pixel art, hand drawn, grain, infantile sketch, 2D art poster, handmade 3D, hand drawn outline, engraving color, 2D art poster 2
08
Upload reference images to enforce a custom visual style across outputs
09
Pass hex color codes via the `colors` parameter to lock brand palettes into every generation
10
Inpainting, outpainting, image-to-image, background replacement, background removal, erase region, image vectorization, crisp upscale, creative upscale, and prompt enhancement. All available via API