# Hunyuan Image 3 Instruct AI Image Generator

> Tencent’s instruction-focused image model for detailed prompts and deliberate composition.

Canonical URL: https://allimageai.com/models/hunyuan-image-3-instruct/

Website: All Image AI

Language: English

Hunyuan Image 3 Instruct is Tencent’s instruction-oriented generation endpoint. It combines an internal prompt-expansion step with an explicit guidance control, making it useful when a scene contains several spatial or material constraints that should be reasoned through before rendering.

## Model details

- Developer: Tencent
- Status on All Image AI: Active
- Last verified: 2026-08-25
- Maximum output: 1K
- Modes: Text to Image
- Reference images: not supported
- Resolutions: 1K
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3

## Best for

- Detailed scene instructions
- Scientific and technical illustration concepts
- Complex material descriptions
- Deliberate composition

## How to use

1. Describe spatial relationships explicitly.
2. Keep Guidance near 3.5 first.
3. Compare a fixed seed with Prompt Expansion on and off.
4. Use a cheaper model for broad exploration.

## Settings explained

### Guidance

Controls adherence to the prompt; the documented default is 3.5.

### Prompt Expansion

Runs the model’s internal recaptioning step before generation.

## Prompt examples

### Material macro

> Macro photograph of a peacock feather showing individual barbs, barbules and iridescent microstructure, controlled side light, scientific accuracy

### Complex interior

> Cutaway reading room inside a lighthouse, spiral stair at left, desk facing the sea, storm light through three windows, physically coherent scale

## Limitations

- Text-to-image only here
- Higher compute cost than lightweight models
- 1K output on the verified route

## Pricing and credits

Credits for one image at default quality and mode. Resolution, quality, reference inputs, edit mode and output count can change the total. Check the Generate button before submitting.

| Resolution | Credits / image |
| --- | --- |
| 1K | 13 |

Failed generations cost 0 credits. Catalog availability does not guarantee runtime availability; the workspace reports whether the selected model can run.

[Credit plans](https://allimageai.com/pricing.md)

## Test results

Testing in progress. No verified benchmark scores have been published for this model.

## Commercial use and image handling

Commercial use is governed by Tencent’s Hunyuan licence and the selected service terms.

Uploads are processed by the configured inference service. You can delete a generation from your history.

## Frequently asked questions

### Why is editing not shown?

The edit operation uses a different compute-second price, so it remains separate until its maximum cost can be bounded safely.

### Does Prompt Expansion change the subject?

It should preserve the requested meaning while adding visual detail; disable it for strict literal prompts.

## Release history and sources

January 26, 2026 — Open weights released

This date applies to the event shown. Later API access, weights and availability on All Image AI may have different dates.

- [github.com](https://github.com/Tencent-Hunyuan/HunyuanImage-3.0)

[Event evidence on the timeline](https://allimageai.com/timeline/#event-hunyuan-image-3-instruct)

## Other versions

- [Hunyuan Image 3.0](https://allimageai.com/models/hunyuan-image-3.md): Base model · 2K
- [Hunyuan Image 2.1](https://allimageai.com/models/hunyuan-image-2-1.md): Steps + guidance · 1K