HunyuanImage-3.0 Instruct
The Instruct release adds reasoning, prompt rewriting and image-to-image generation to HunyuanImage-3.0. It supports editing and combining multiple reference images.
Model design and access
The model can reason about the requested composition before generating an image. The card documents at least eight 80 GB GPUs for its reference Instruct configuration. The downloadable model is separate from the hosted hy-image-v3 service and its scheduled API retirement.HunyuanImage-3.0 Instruct official model documentation
Practical use
Practical guidance. Review whether an edit preserves details that were not part of the instruction. For multiple references, state which image owns identity, layout and style. Save the expanded prompt when the model rewrites the request.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | Downloadable |
| License | Tencent Hunyuan Community License; excludes the EU, UK and South KoreaHunyuanImage 3.0 community license |
| Parameters | 80B total; 13B activeHunyuanImage-3.0 Instruct official model documentation |
Related models
- HunyuanImage-3.0
Tencent · HunyuanImage
An autoregressive text-to-image model with 80B total and 13B active parameters.
Launch date not confirmedDownloadable - HunyuanImage-3.0 Instruct Distil
Tencent · HunyuanImage
HunyuanImage-3.0 Instruct Distil keeps the Instruct family’s image generation, editing and prompt reasoning workflow while reducing the sampling schedule.
Launch date not confirmedDownloadable