Search papers, labs, and topics across Lattice.
The authors introduce HunyuanImage 3.0, a multimodal model unifying understanding and generation within an autoregressive framework, with a focus on image generation. They achieved this by using meticulous data curation, advanced architecture design, a native Chain-of-Thoughts schema, progressive pre-training, aggressive post-training, and efficient infrastructure. The resulting Mixture-of-Experts (MoE) model, with over 80 billion parameters (13B active per token), demonstrates state-of-the-art performance in text-image alignment and visual quality, rivaling previous models.
The largest open-source image generative model to date, HunyuanImage 3.0, achieves state-of-the-art performance using a Mixture-of-Experts architecture and native Chain-of-Thoughts schema.
We present HunyuanImage 3.0, a native multimodal model that unifies multimodal understanding and generation within an autoregressive framework, with its image generation module publicly available. The achievement of HunyuanImage 3.0 relies on several key components, including meticulous data curation, advanced architecture design, a native Chain-of-Thoughts schema, progressive model pre-training, aggressive model post-training, and an efficient infrastructure that enables large-scale training and inference. With these advancements, we successfully trained a Mixture-of-Experts (MoE) model comprising over 80 billion parameters in total, with 13 billion parameters activated per token during inference, making it the largest and most powerful open-source image generative model to date. We conducted extensive experiments and the results of automatic and human evaluation of text-image alignment and visual quality demonstrate that HunyuanImage 3.0 rivals previous state-of-the-art models. By releasing the code and weights of HunyuanImage 3.0, we aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant multimodal ecosystem. All open source assets are publicly available at https://github.com/Tencent-Hunyuan/HunyuanImage-3.0
