Canva Craft
Canva Craft is Canva's image-generation model for design-oriented visuals from text prompts. Available in Dream Lab and through Canva's image-generation API, it creates images up to 2K resolution in multiple aspect ratios. Its two-stage process plans composition before rendering, supporting positioned objects, visual hierarchy, and embedded text. Use cases include posters, social posts, presentations, infographics, advertising, editorial imagery, portraits, landscapes, illustration, and conceptual artwork. Prompts can be written in any language, and generated images can contain text in 55 languages. Within Canva, Craft outputs can be converted into editable design elements with Magic Layers. Developers can request Craft through Canva's authenticated API, with image-generation jobs consuming the requesting user's AI credits. Canva applies input and output safety checks.
Learn more about Canva Craft
Seedream 4.0
Seedream 4.0 is a next-generation multimodal AI image generation and editing model that unifies text-to-image creation and text-guided image editing within a single architecture, delivering professional-grade visuals up to 4K resolution with exceptional fidelity and speed. It’s built around an efficient diffusion transformer and variational autoencoder design that lets it interpret text prompts and reference images to produce highly detailed, consistent outputs while handling complex semantics, lighting, and structure reliably, and it offers batch generation, multi-reference support, and precise control over edits such as style, background, or object changes without degrading the rest of the scene. Seedream 4.0 demonstrates industry-leading prompt understanding, aesthetic quality, and structural stability across generation and editing tasks, outperforming earlier versions and rival models in benchmarks for prompt adherence and visual coherence.
Learn more about Seedream 4.0
ERNIE-Image
ERNIE-Image is an open text-to-image generation model developed by Baidu, designed to deliver high-quality visuals with strong instruction accuracy and controllability. It is built on a single-stream Diffusion Transformer (DiT) architecture with around 8 billion parameters, allowing it to achieve state-of-the-art performance among open-weight image models while remaining relatively efficient. The model includes a built-in prompt enhancement system that expands simple user inputs into richer, structured descriptions, improving the quality and consistency of generated images. ERNIE-Image is optimized for complex instruction following, enabling accurate rendering of text within images, structured layouts, and multi-element compositions, making it particularly suitable for use cases like posters, comics, and multi-panel designs. It supports multilingual prompts, including English, Chinese, and Japanese, broadening accessibility and usability across regions.
Learn more about ERNIE-Image
Qwen-Image-2.0
Qwen-Image 2.0 is the latest AI image generation and editing model in the Qwen family that combines both generation and editing in a single unified architecture, delivering high-quality visuals with professional-grade typography and layout capabilities directly from natural-language prompts. It supports text-to-image and image editing workflows with a lightweight 7 billion-parameter model that runs quickly while producing native 2048x2048 resolution outputs and handling long, detailed instructions up to about 1,000 tokens so creators can generate complex infographics, posters, slides, comics, and photorealistic scenes with accurate, well-rendered English and other language text embedded in the visuals. The unified model design means users don’t need separate tools for creating and modifying images, making it easier to iterate on ideas and refine compositions.
Learn more about Qwen-Image-2.0