China first open-source AI image generation model HiDream-I1 released

Recently, a major announcement emerged in the field of artificial intelligence: the open-source image generation model HiDream-I1 with 1.7 billion parameters has been officially released. The model not only excels in generating high-quality images but also surpasses the highly acclaimed GPT-4o in certain aspects.

Developed by the Chinese company Zhixiang Future, the HiDream-I1 model successfully made it onto the artificial intelligence benchmarking platform rankings, quickly becoming a new benchmark for open-source models. Within 24 hours of its release, it topped the list, becoming the first Chinese independently developed generative AI model to achieve this feat.

Intelligent Elephant Future plans to release another AI model called HiDream-E1 in the near future. This AI model will support interactive image editing, allowing users to freely modify the style and content of images, further enhancing the flexibility and practicality of AI image generation.

Key Features

• Superior Image Quality – Produces exceptional results across multiple styles including photorealistic, cartoon, artistic, and more. Achieves state-of-the-art HPS v2.1 score, which aligns with human preferences.
• Best-in-Class Prompt Following – Achieves industry-leading scores on GenEval and DPG benchmarks, outperforming all other open-source models.
• Open Source – Released under the MIT license to foster scientific advancement and enable creative innovation.
• Commercial-Friendly – Generated images can be freely used for personal projects, scientific research, and commercial applications.

More

HiDream-I1 Model: https://huggingface.co/HiDream-ai/HiDream-I1-Full

HiDream-I1 Code: https://github.com/HiDream-ai/HiDream-I1