Core Models
Our core AI models are considered non-generative AI models that allow for enhancements across a variety of images and files. These AI models are available to process locally on your computer. Cloud rendering is not available.
Denoising and Sharpening only apply with regards to the upscaling process. For dedicated Denoise and Sharpen tools, look to Topaz Photo.
Standard
Standard 2 is the general-purpose upscaling model recommended for most images and graphics. It delivers clean, reliable enlargement across a wide range of image types while maintaining true-to-input detail and structure.
- Trusted, industry-standard image upscaling
- Preserves natural detail and texture
- Balanced model is ideal default choice for photos, graphics, and print
As an alternative for smaller images, try Standard Max.
High Fidelity
High Fidelity 3 is a precision upscaling model designed to deliver the best results on high-resolution sources. This model improves on High Fidelity 2 with enhanced resolution scaling and minimal detail loss or artifacting. It focuses on preserving original image structure and realism, reducing the plasticky appearance and excess artifacts seen in earlier models while maintaining a clean, natural output.
High Fidelity 3 performs best on medium to high and high quality images, including modern photos, RAW images, and smartphone photography, and supports a wide range of content such as portraits, landscapes, wildlife, and images with graphics.
- Higher quality upscaling with improved detail retention and fewer artifacts compared to High Fidelity 2.
- Preserves natural image structure with reduced plasticky appearance and more consistent textures.
- Optimized for high-quality images, delivering clean results beyond traditional resizing.
For lower resolution or lower quality images, Wonder is a better starting point.
High Fidelity 2 is a precision upscaling model designed for images that are already high resolution and well captured. It focuses on enlarging images while preserving fine detail, natural texture, and structural accuracy, avoiding aggressive reconstruction or over-sharpening.
- High-fidelity upscaling for already high-resolution sources
- Preserves fine detail and original texture during enlargement
To keep the look of grain or noise in the image use High Fidelity v3.
Low Resolution
Low Resolution 2 is optimized for images that start with limited detail, heavy compression, or small dimensions. It focuses on intelligent reconstruction to recover usable detail from challenging sources without modifying the source characteristics.
- Designed for low-resolution and compressed images
- Recovers usable detail from difficult inputs
- Improves clarity while avoiding exaggerated artifacts
If false resolution is a problem, try Recover 2 or Wonder 3.
Text & Shapes
Text & Shapes is designed for images dominated by typography, logos, UI elements, and geometric shapes. It prioritizes edge clarity and shape accuracy during upscaling.
- Preserves sharp edges and geometric structure
- Ideal for logos, UI, and graphic design assets
- Maintains legibility during enlargement
Art & CG
Art & CGI is tailored for non-photographic imagery such as illustrations, digital art, renders, and stylized graphics. It preserves clean edges, flat color regions, and intentional line work without introducing photographic texture.
- Optimized for illustrations, digital art, and CGI
- Preserves clean edges and flat color areas
- Avoids unwanted photographic texture or noise
For AI generated images, try using Redefine to add creative distinction.
Generative Models
