HunyuanImage-3.0

model paper
Tencent
Native multimodal autoregressive model for image generation, built on the Hunyuan-A13B backbone (80B+ total, 13B active). Largest open-source image generative model at time of release. Capable of instant updates from text, sketch, or voice.
GitHubHuggingFacePaper (arXiv)

Outputs 2

HunyuanImage-3.0

model
GitHubHuggingFace
Architecture MOE
Active params 13B

HunyuanImage 3.0 Technical Report

paper

Technical report for the native multimodal autoregressive image generation model built on Hunyuan-A13B.

Paper (arXiv)

arXiv: 2509.23951

generationvisionmultimodalopen-weight

Related