Pricing for Developers


Output Shape Size default: 768x768

Pricing will be calculated by using the closest dimension


steps default: 20 ( $0.000598 every 5 steps above 20 )

Cost Per API Call

$ 0.000000

For enterprise pricing and custom weights or models

contact our sales team.

Kandinsky 2.2

Kandinsky 2.2, a groundbreaking advancement over its predecessor, Kandinsky 2.1. With the integration of the powerful CLIP-ViT-G image encoder and the innovative ControlNet support, Kandinsky 2.2 is set to redefine the boundaries of aesthetic image creation and text comprehension.

At the heart of Kandinsky 2.2 lies the state-of-the-art CLIP-ViT-G image encoder, a transformative addition that amplifies the model's ability to craft visually stunning images while enhancing its text understanding capabilities. Complementing this is the ControlNet mechanism, a strategic inclusion designed to offer users unparalleled control over the image generation process.

Advantages

  1. Enhanced Image Aesthetics: The CLIP-ViT-G encoder ensures the generation of visually richer and more captivating images.

  2. Superior Text Understanding: With the new encoder, the model boasts an improved comprehension of text, bridging the gap between textual prompts and visual outputs.

  3. Precision Control: The ControlNet support empowers users to guide the image generation process, ensuring outputs that align with their vision.

  4. Optimized Performance: The combined power of CLIP-ViT-G and ControlNet results in a significant boost in the model's overall performance.

Use Cases

  1. Digital Art Creation: Artists can harness Kandinsky 2.2 to craft digital artworks that resonate with depth and detail.

  2. Content Generation: Ideal for content creators seeking to generate visuals based on textual prompts or narratives.

  3. Interactive Design: Designers can iteratively shape their designs, making real-time adjustments guided by text.

  4. Educational Tools: Can be integrated into learning platforms, allowing students to explore the interplay between text and visuals.

  5. Gaming and AR: Enhance user immersion in games or AR experiences by generating visuals based on in-game narratives or user prompts.

Kandinsky 2.2 License

Kandinsky 2.2's permissive license ensures that users, be they individual creators, businesses, or developers, can utilize the model for a myriad of commercial purposes without the constraints typically associated with restrictive licenses.