Topaz Image Generative

Topaz Image Generative

Synthesizes detail the source never had. Every engine here takes an optional prompt plus creativity and texture controls, so you can steer what gets invented or leave it to Topaz.

Topaz Image Generative API on MachGen. Compare supported generation modes, examples, and current pricing from the live pricebook.

Why choose Topaz Image Generative

Key strengths

• Synthesizes detail the source never contained, rather than enlarging what is there.
• Every engine takes an optional prompt to steer what gets invented; leave it empty to let Topaz decide.
• Recovers usable images from sources other upscalers cannot help.

Good fit for

• Restoring old, damaged, or heavily compressed photographs.
• Adding believable texture to a soft or low-detail source.
• Creative reinterpretation where faithfulness is not the goal.

Engines in Topaz Image Generative

This model is a price tier: every engine below costs the same, so the choice is about the source, not the budget. Pick one in Generate under the model; leaving it unset uses the first.

• Redefine - The most creative freedom, and the only image engine steered by a text prompt. Creativity and texture control how far it departs from the source.
• Wonder 3 - The newest Wonder model, with strong detail recovery at the lowest vendor cost in this tier. Takes no prompt.
• Wonder 2 - Adds pre-downscale and colour correction, useful when the source needs cleaning up as well as enlarging.
• Standard MAX - The generative counterpart to Standard V2, for sources up to 24 MP.
• Recover 3 - Built for heavily degraded photographs, with an enhancement-strength control to dial how aggressively it rebuilds. Sources up to 24 MP.

Inputs and output settings

Upscaling is live in Models, Generate, and on a completed generation's TaskCard.

Using Topaz Image Generative

1. Upload a source image or choose one from Assets in Models or Generate. You can also start from a completed image's TaskCard.
2. Pick an engine matched to the source, then a scale factor.
3. Review the estimate, then submit the job.
4. Inspect the enhanced image when processing completes.

How to get better Topaz Image Generative results

• Detail is invented - do not use it where the image must remain accurate.
• With Redefine, describe what the subject is; creativity controls how far it may depart from the source.
• Check faces and text carefully, since those are where synthesis is most visible.

Limits and availability

• Pick a scale factor, not a size: the server reads the source and derives the output, clamped to this model's ceiling.
• Billing is per output megapixel, so the factor you choose sets the price.
• Direct uploads are limited to 32 MiB; generated MachGen images can be selected from Assets or a completed TaskCard.
• Current rates come from the live MachGen pricebook.
• Topaz Image Generative is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.

Learn more

For model background, technical details, and architecture, see the official Topaz Wonder documentation.

Use the MachGen Playground to upload or choose a source image. You can also open a completed image's TaskCard and choose Upscale.

Official documentation may describe capabilities or parameters that are not currently exposed by MachGen. Use the Inputs and output settings section above as the source of truth for this page.

Topaz Image Generative pricing

See Pricing for currently published MachGen rates. Models, Generate, and the completed video's TaskCard show the current estimate before an Upscale job is submitted.

Topaz Image Generative on MachGen

Deployment: Pass-through. Topaz Image Generative is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.

Upload or choose a source in Models or Generate, or open a completed MachGen generation's TaskCard and choose Upscale.