Tencent has upgraded and open-sourced Hunyuan, its text-to-image large model (LM).
The model is claimed to be the first Chinese diffusion model based on Transformer (DiT) architecture. It also has a parameter size of 1.5 billion and is a text-to-image open-source model.
The company claims that the upgraded model provides improved efficacy in generating visual models, and has demonstrated an improvement of over 20% compared to its previous versions.
The model is accessible on the Hugging Face platform and Github, and is free for commercial use.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.