Recraft 20B — Design-Native Text-to-Image
Recraft 20B is a large-scale text-to-image model that “thinks in design language.” It is tuned for layouts, typography, brand-safe compositions, and clean text rendering, making it ideal for social graphics, ads, presentations, and product visuals where design quality matters as much as realism.
Why it stands out
-
Design-first image generation
Strong understanding of layout, hierarchy, and composition for poster-style images, thumbnails, and marketing assets.
-
Sharp, readable text on images
Generates high-quality embedded text—from short labels to longer headlines—directly inside the image.
-
Rich style system
Curated style presets (for example, igital_illustration/grain make it easy to switch between illustration, painterly, and graphic styles without rewriting prompts.
-
Brand and identity friendly
Handles logos, icons, and UI-like designs with improved anatomy, alignment, and visual consistency.
-
Vector-aware design
Originates from a system that supports both raster and vector workflows, making it particularly strong for flat design, icons, and graphic illustration.
Key capabilities
- Advanced prompt understanding for detailed scenes, characters, and layouts.
- Stable composition suitable for posters, covers, landing-page hero sections, and ad creatives.
- Improved anatomy and perspective compared to many generic art models.
- Good text–image integration so titles, slogans, and labels feel part of the design rather than pasted on.
Controls and parameters
- prompt – natural-language description of the scene, style, layout, and embedded text you want.
- aspect_ratio – choose from standard ratios (for example 1:1, 16:9, 9:16) to match feeds, banners, and story formats.
- style – select from model-defined style families such as igital_illustration/grainto quickly shift aesthetic direction.
- enable_base64_output – when enabled via API, returns the image as a base64 string instead of a URL (useful for certain integrations).
Outputs are delivered as high-quality raster images suitable for web, slides, and print-oriented workflows.
Pricing
Simple per-image billing:
- $0.022 per generated image
How to use
- Enter a prompt describing subject, composition, design style, and any on-image text (titles, slogans, labels).
- Choose an aspect_ratio that matches the target placement (for example, 16:9 for banners, 9:16 for vertical stories).
- Select a style preset that fits your use case (illustration, grainy poster look, etc.).
- (Optional) Enable base64 output if your integration needs inline image data instead of URLs.
- Run the job and download or embed the generated image from the response or dashboard.
Pro tips
- Be explicit about layout (e.g., “centered character with headline at the top, small subtitle below”) to leverage Recraft’s design-native strengths.
- For text-heavy designs, spell out exact wording and, if needed, describe typography (for example, “bold sans-serif headline, clean minimal layout”).
- Use aspect ratios that match the final placement so you avoid cropping away important elements.
- When exploring brand visuals, keep the core prompt stable and tweak style or minor wording to generate consistent variations.