Header Logo

Docs

Start

Header Logo

Start

Header Logo

Docs

Start

All Models

WaveSpeed logo

Hunyuan Image 3 Instruct

Hunyuan Image 3 Instruct generates images from text and revises supplied images through written instructions, supporting concept creation and edits guided by visual context.

Image editing

Price per run for each listed option.

OfferingPrice (USD)
All listed options$0.45 / run
  • Images count: 1, 2 (same price).

Image generation

Price per run for each listed option.

OfferingPrice (USD)
Per run$0.45 / run

Hunyuan Image 3 Instruct

Hunyuan Image 3 Instruct combines image generation from text with instruction-based editing of supplied images. For generation, a written description defines the desired image. For editing, source imagery establishes the visual context and the accompanying text describes what should change.

The generation task accepts descriptions in Chinese or English. Editing takes scene structure and object relationships into account, with changes such as replacing backgrounds, removing objects, or adjusting product colors. The model is designed to retain the source image's style and composition during revisions.

What it's good for

The paired tasks suit creative work that moves from a brief into image development and revision. A team can generate a campaign concept, then edit the chosen artwork to explore a different setting. Product teams can revise existing photography, while illustrators can develop variations on established artwork. Reference images provide a defined starting point for changes that need to relate to an existing visual.

Model ID

wavespeed-ai/hunyuan-image-3-instruct

Output Type

Image

Pricing

$0.45 / run

Provider

WaveSpeed

Hosting

Passthru

Hunyuan Image 3 Instruct

Hunyuan Image 3 Instruct combines image generation from text with instruction-based editing of supplied images. For generation, a written description defines the desired image. For editing, source imagery establishes the visual context and the accompanying text describes what should change.

The generation task accepts descriptions in Chinese or English. Editing takes scene structure and object relationships into account, with changes such as replacing backgrounds, removing objects, or adjusting product colors. The model is designed to retain the source image's style and composition during revisions.

What it's good for

The paired tasks suit creative work that moves from a brief into image development and revision. A team can generate a campaign concept, then edit the chosen artwork to explore a different setting. Product teams can revise existing photography, while illustrators can develop variations on established artwork. Reference images provide a defined starting point for changes that need to relate to an existing visual.