Captured photo
Kandinskiy 2.2
60 +

Gallery

Robotic storefront

Original prompt
Full prompt

Photorealistic architectural street photograph of an independent robot shop at blue hour. Create a striking MODERN storefront title design that reads exactly "ropewalk". Invent expressive, memorable lettering and integrate it thoughtfully into the architecture; the title should feel like an original design statement and remain clearly readable. Explore the typography, sign materials, lighting and facade creatively. A graphite-and-mint visual identity, intriguing robots visible through the shop window, a welcoming interior, believable urban details and subtle reflections on the pavement. Confident composition showing both the title and the character of the shop. Square image, no watermark.

Photorealistic architectural street photograph of an independent robot shop at blue hour. Create a striking MODERN storefront title design that reads exactly "ropewalk". Invent expressive, memorable lettering and integrate it thoughtfully into the architecture; the title should feel like an original design statement and remain clearly readable. Explore the typography, sign materials, lighting and facade creatively. A graphite-and-mint visual identity, intriguing robots visible through the shop window, a welcoming interior, believable urban details and subtle reflections on the pavement. Confident composition showing both the title and the character of the shop. Square image, no watermark.

Glass beetle

Original prompt
Full prompt

A translucent mint-glass mechanical beetle resting on black volcanic stone. Fine wires and tiny moving components visible inside its body, iridescent wings slightly open. Extreme macro photography, hard studio lighting, deep shadows, startling physical realism.

A translucent mint-glass mechanical beetle resting on black volcanic stone. Fine wires and tiny moving components visible inside its body, iridescent wings slightly open. Extreme macro photography, hard studio lighting, deep shadows, startling physical realism.

Street-fashion portrait

Original prompt
Full prompt

Editorial street-fashion portrait of a young adult Japanese woman in Tokyo, with the inventive, eclectic spirit of early-2000s Harajuku street style. Playful experimentation, unexpected combinations, expressive layering, bold color relationships and a strong sense of personal identity. Create an original, coherent look; choose her clothing, hairstyle, accessories and pose freely within that aesthetic. Full-length portrait with a strong composition and a natural, self-assured presence. Eye-level 35mm street photography, direct flash mixed with daylight, subtle film grain, honest skin texture and an informal documentary feel. Square image, no magazine layout or added captions.

Editorial street-fashion portrait of a young adult Japanese woman in Tokyo, with the inventive, eclectic spirit of early-2000s Harajuku street style. Playful experimentation, unexpected combinations, expressive layering, bold color relationships and a strong sense of personal identity. Create an original, coherent look; choose her clothing, hairstyle, accessories and pose freely within that aesthetic. Full-length portrait with a strong composition and a natural, self-assured presence. Eye-level 35mm street photography, direct flash mixed with daylight, subtle film grain, honest skin texture and an informal documentary feel. Square image, no magazine layout or added captions.

Percs

Fine control
High quality

About

Kandinsky 2.2 is an open-source text-to-image model that produces high-resolution, photorealistic and stylized images from natural language prompts while also supporting guided image editing and blending. It delivers detailed textures, realistic lighting and fine detail up to 1024×1024, making it suitable for portraits, landscapes, cityscapes, fantasy concepts and abstract art. Improved text–image alignment helps the model follow complex instructions and yield consistent results across varied prompts. Beyond standard text-to-image generation, Kandinsky 2.2 offers text-guided image manipulation: you can inpaint, extend, or alter parts of an existing photo by describing changes in words. The model also supports image interpolation and morphing between sources, letting you create hybrid visuals or smooth transitions guided by text. Integration with ControlNet gives precise control over composition—condition generation on sketches, edge maps or other structural inputs—to produce repeatable, layout-driven outputs that are valuable for concept artists, illustrators and designers. Practically, Kandinsky 2.2 is useful for rapid prototyping in advertising, game and film concepting, portrait and landscape generation, and image restoration or creative recomposition. Its open-source, permissive licensing enables customization and integration into research and commercial pipelines. Note that high-quality and high-resolution outputs require significant GPU resources, and the model can reflect dataset biases or occasionally produce artifacts. Overall, Kandinsky 2.2 balances realism, controllability and versatility, offering creatives and researchers a powerful tool for generating and editing images from text prompts.
Release date:: 06.10.2026
WebsiteLicenseGithub

Settings

Negative Prompt-  Things you don't want to see in the output
Width-  Width of output image. Lower the setting if hits memory limits.
Height-  Height of output image. Lower the setting if hits memory limits.
Inference Steps-  Number of denoising steps
Inference Steps Prior-  Number of inference steps for prior