Skip to content
Hunyuan image generator

Build a scene from words with Hunyuan Image V3.

Describe an image you want to create and explore it with Hunyuan Image V3. Set the composition through your prompt and aspect-ratio choice. Refine the wording after inspecting the generated result.

Compare plans and credits
Illustration of a brick library beside a tree-lined square with warm evening windows
Illustrative image made with Nano Banana 2.1 in Sceneform, not with Hunyuan Image V3.

What you can create

Describe the scene in text

Name the subject, materials, surroundings, and lighting. This model generates images from text without an uploaded reference-image input.

Choose a frame for the subject

Select a square, landscape, or portrait aspect ratio. Explain where the subject should sit within that frame.

Review advanced generation settings

The model exposes a negative prompt, seed, guidance scale, and inference-step controls. Change one setting at a time when comparing results.

From your idea to finished content

  1. Open Hunyuan Image V3

    Use the button to select the image model. Prepare a focused text description for the scene you want.

  2. Set the scene and frame

    Describe the composition and choose an aspect ratio. Add a negative prompt if you want to discourage particular elements.

  3. Generate and refine

    Review the settings and cost, then generate. Check the scene against your brief before changing the wording or advanced controls.

Prompts to start with

Use these briefs as a starting point. Replace the details with your own product, audience, and style.

An architectural illustration

A small brick library stands beside a tree-lined square. Show the building from street level with warm evening windows and soft shadows.

A material-focused still life

A smooth river stone rests beside a folded indigo cloth. Use soft side lighting and a simple background with visible fabric texture.

Frequently asked questions

Does this page generate images or videos?
It opens Hunyuan Image V3 for image generation. It does not provide a Hunyuan video or 3D-model workflow.
Can I upload a photo as a reference?
This model's verified workflow accepts text only. Choose an image-reference model when your project depends on an existing photograph.
What should I put in the negative prompt?
List elements you want the model to avoid. Treat these as instructions rather than guaranteed exclusions, and inspect the result.
Can I choose portrait and landscape outputs?
Yes. The model offers square, portrait, and landscape aspect ratios. Choose the shape before generating.

Create your next project

Start with a clear brief. Review the result. Make it yours.