Image Upscale increases image resolution up to 10x using AI models that add genuine detail, not just interpolation. Choose the right model for your use case — fast general-purpose upscaling, seamless tiling for textures, professional-grade with face enhancement and text refinement, or noise and compression cleanup.
Whether you need a small product thumbnail blown up for print, a noisy photo cleaned up and sharpened, or a game texture tiled at 4x resolution, there's a model designed for that specific task. Batch mode processes up to 10 images in parallel.
What you can do
- upscale_image — upscale a single image up to 10x with model selection and quality controls
- batch_upscale — process up to 10 images in parallel, with per-image model overrides
- list_models — browse all available models with pricing, maximum scale factors, and variants
- Choose a target resolution (720p to 4K) instead of a scale factor when you have a specific output size in mind
Who it's for
Photographers and retouchers restoring old or low-resolution images, game developers upscaling texture assets, e-commerce teams enlarging product photos for high-DPI displays, and anyone who needs a small image at a larger size without visible degradation.
How to use it
- Use upscale_image with your image URL — the default model handles most general photos well at 2x
- Set target_resolution if you need a specific output size like 4K instead of a scale multiple
- Use the topaz model for professional work with face enhancement, text sharpening, or CGI upscaling
- Use batch_upscale to process multiple images in one call, mixing models per image if needed
Getting started
Works immediately with no setup — pass any public image URL to upscale it.
