Image Models

Topaz Image Web — Models

Topaz Image Web image models are organized by family based on their primary function. Each family is designed to solve a different type of image enhancement challenge.

Precision Upscaling

Precision upscaling models designed to enlarge images while preserving original source characteristics, detail, and structure.

Upscale Standard

General-Purpose Upscaling

The recommended starting point for most upscaling projects. Delivers clean, reliable enlargement across a wide range of image types while maintaining true-to-input detail and structure. Ideal for photos, graphics, and print.

Upscale High Fidelity

High-Fidelity Upscaling for Professional Photos

Precision upscaling designed for images that are already high resolution and well captured. Focuses on enlarging images while preserving fine detail, natural texture, and structural accuracy. Best used when the source image quality is strong and the goal is clean, accurate enlargement.

High Fidelity V3

Improved Fidelity Upscaling with Advanced Recovery

An improvement on Upscale High Fidelity with enhanced resolution scaling and minimal detail loss or artifacting. Reduces plasticky appearance and excess artifacts while maintaining a clean, natural output. Performs best on medium to high quality images including modern photos, RAW images, and smartphone photography.

Low Resolution

Recovery-Focused Upscaling for Low Quality Sources

Optimized for images that start with limited detail, heavy compression, or small dimensions. Focuses on intelligent reconstruction to recover usable detail from challenging sources without modifying the source characteristics.

Upscale CGI

Precision Upscaling for Artwork and Computer Graphics

Tailored for non-photographic imagery such as illustrations, digital art, renders, and stylized graphics. Preserves clean edges, flat color regions, and intentional line work without introducing photographic texture.

Wonder Upscale

Generative Upscaling for Low Quality Images

A generative upscaling model designed to repair and improve small to medium sized low-resolution images. Intelligently reconstructs missing detail while prioritizing realism and avoiding unnecessary changes to objects, textures, or overall image structure.

Natural Enhance

Natural Detail Enhancement for Faces, Fur, and Fabric

Enhances images with natural and realistic detail, particularly effective on faces, fur, and fabric. Offers adjustable enhancement strength for control over how much detail is added.

Region Enhance

Selective Enhancement for Specific Areas

Allows targeted enhancement of specific regions within an image using a mask. Only masked regions receive full enhancement while other areas receive lighter processing. Useful for workflows where selective control is needed.

Bloom — Creative Upscaling

Creative upscaling models designed for GenAI images and artwork, adding new detail, texture, and visual elements beyond traditional enhancement.

Bloom

Creative Upscaling for AI-Generated Images

A generative upscaling model built for native GenAI images. Enlarges AI-generated artwork, illustrations, and CG renders by up to 6× while creatively introducing new detail, texture, and visual elements. Supports optional prompt input and adjustable creativity levels ranging from subtle to bold.

Bloom Realism

Realistic Detail Enhancement for AI-Generated Images

Removes the plastic, artificial feel of AI-generated images by restoring natural detail and texture across skin, hair, fabrics, metals, and other materials. Upscales native GenAI images by up to 6× while producing more grounded, believable results.

Bloom Precision

High-Precision Creative Upscaling

A precision-focused variant within the Bloom family. Delivers high quality upscaling with optional face enhancement and color correction controls. Supports scale factors up to 6×.

Redefine

Generative Diffusion Upscaling with Prompt-Based Control

A generative image enhancement model that adds new detail and rebuilds structure when original image information is missing or degraded. The only local, prompt-driven generative image model available. Supports image prompts, creativity controls, and region-preserving masks for targeted enhancement. Especially valuable for CGI, VFX, and creative workflows.

Wonder 3

Advanced Generative Realism for All Image Types

A generative upscaling model delivering higher quality results with more natural detail and improved realism across a wide range of image types. Handles both high-quality inputs and heavily degraded or compressed images. Supports Low, Medium, and High enhancement levels for greater control. Performs well on portraits, landscapes, wildlife, and complex textures such as fabrics, metallics, and skin.

Restore — Sharpen

Specialized sharpening models designed to reduce blur and improve clarity across different types of blur and source conditions.

Sharpen Standard

General-Purpose Sharpening

An all-purpose sharpening model designed for images with mild lens softness or slight motion blur. Recommended starting point for most photography sharpening workflows. Produces clear images with crisp detail without darkening or brightening pixel edges.

Sharpen Strong

Severe Blur Recovery

Applies maximum reconstruction strength for severely blurred or out-of-focus images where standard sharpening is insufficient. Best used selectively to avoid artifacts.

Auto Sharpen

Intelligent Automatic Sharpening

Automatically selects the best sharpening approach for the image using intelligent model selection. Best used when you want a fast, one-click result without manually choosing a sharpening model.

Lens Blur

Defocus and Lens Blur Correction

Designed to correct optical defocus and lens-related blur. Targets defocus and lens imperfections rather than motion blur. Delivers consistent sharpening across heavily blurred regions.

Sharpen Motion Blur

Motion Artifact Correction

Optimized to correct directional streaking and smearing caused by camera shake or subject movement during exposure. Best suited for action or handheld captures.

Sharpen Standard Portrait

Detail Enhancement for People

Trained to enhance fine detail in images of people, prioritizing natural skin texture and facial structure. Avoids harsh edges and unnatural texture. Best suited for portraits and close-up human subjects.

Sharpen Standard Wildlife

Detail Recovery for Nature Imagery

Trained to enhance fine detail in wildlife and nature images while preserving organic textures. Optimized for feathers, fur, scales, and natural patterns. Designed for high-resolution nature photography.

Restore — Denoise

Specialized noise reduction models for reducing noise and grain while preserving detail.

Denoise Normal

General-Purpose Noise Reduction

Designed for everyday images with low to medium noise caused by low-light conditions or compression. Preserves fine detail and natural texture. Recommended starting point for most denoising workflows.

Denoise Strong

High-Noise Detail Preservation

Targets medium to high noise levels from very low-light or high ISO captures. Balances stronger noise reduction with careful detail retention. Use when Denoise Normal does not sufficiently clean the image.

Denoise Max

Highest-Quality Denoising with Intelligent Detail Recovery

A next-generation generative denoising model delivering the highest quality results in the Denoise family. Removes noise and grain with significantly reduced artifacts while intelligently sharpening blurry areas without affecting already sharp detail. Performs best on medium to high quality images with light noise or structured grain.

Restore — Faces

Specialized models for recovering and enhancing facial detail.

Face Recovery Natural

Natural Face Recovery with Reference Support

Recovers and enhances facial detail with a natural, realistic output. Requires a reference image to guide the recovery. Best suited for portraits where a source reference is available.

Face Detailer

Detail Enhancement for Faces

Enhances facial features with natural detail in skin, eyes, and hair. Reduces plastic or synthetic artifacts for more realistic, lifelike results. Adds realistic texture to skin surfaces.

Face Recovery v3 Alpha

Next-Generation Face Recovery

An advanced face recovery model offering separate controls for enhancement strength and recovery strength. Requires a reference image. Currently in alpha — results may vary depending on source quality.

Restore — Cleanup

Models designed to remove unwanted objects, artifacts, and surface damage.

Dust Scratch

Automatic Blemish Cleanup

Automatically detects and removes dust, specks, and small blemishes across the entire image in a single pass. Designed for film scans, archival photos, and damaged prints.

Text Recovery

Restore Text in Images

Specialized model for recovering and sharpening text in images that have been compressed, degraded, or downscaled. Supports scale factors up to 8× with adjustable deblur, denoise, and decompression controls.

Background Removal

Remove Backgrounds from Images

A high-quality image segmentation model that isolates subjects from their background with precise, production-ready edge handling. Preserves fine details such as hair, soft edges, and semi-transparent elements. Outputs both transparent background images and reusable masks for downstream workflows.

Color & Lighting

Models designed to correct color balance, exposure, and lighting.

Colorizer

Colorize Black and White Images

Automatically colorizes black and white and faded images into natural, full-color results using context-aware color mapping. Intelligently assigns realistic hues based on scene content. Ideal for restoring archival photos and historical imagery.

Relight

Intelligent Lighting and Color Correction

Automatically adjusts lighting, exposure, highlights, and shadows to produce balanced, well-lit images. Offers manual controls for fine-tuning lighting, highlight, and shadow levels.