LiblibAI completed three rounds of financing within one year, and the ARR exceeded $1.5 million in half a year of commercialization

A few days ago, "LiblibAI" (hereinafter referred to as "LiblibAI") announced that it had completed three rounds of financing within one year, with a total amount of hundreds of millions of yuan. Source Code Capital, Gaorong Ventures and GSR Ventures participated in the angel round of investment, with the second round led by strategic investors and the latest round led by Mingshi Capital. Foresight Capital acted as the exclusive financial advisor for multiple rounds of financing. This round of funds will be mainly used in three aspects: building a large-scale computing power middle platform to support the demand for massive user inference and training computing power; R&D of plug-ins, fine-tuning models and control capabilities based on image models; Support and operation of the developer ecosystem and the authors of the original model.

LiblibAI's financing journey
Date of disclosure	The amount of the transaction	Funding rounds	Investors
2024-07-29	Not disclosed	Series A	Mingshi Capital
2024-01-01	Not disclosed	Pre-A round	Strategic investor
2023-07-24	Not disclosed	Angel wheel	Source Code Capital Gao Rong Venture Capital GSR Ventures

Founded in May 2023, LiblibAI is an AI image generation platform, dedicated to the creation and sharing of AI content, and completely changing the original creation methods of designers, painters, and self-media creators.

So far, the platform has accumulated nearly 10 million professional AI image creators, produced and shared more than 230 million AI images, including illustration, animation, design, photography and other forms of visual art expression. The number of original models in the community exceeds 100,000, and a complete ecological chain covering AI content creation, sharing, copyright, and sales has been built, providing "artificial intelligence +" productivity tools in the new era for people in the image industry.

The AI painting models on LiblibAI include LoRA, Checkpoint, Textual Inversion, Hypernetwork, etc., as well as more complex Comfy UI workflow templates and tutorials. Users can download these models or use the online Stable Diffusion to create the AI drawings they need.

To use their online tools, users need to use the hash points in their account. Activate the website membership to increase the daily computing power points, and get the benefits of increasing cloud storage space, accelerated computing, and using members-exclusive models. Some models need to be used after purchasing a membership, and a large number of application models such as design, photography, and architecture are launched in the member-only models to support commercial purposes.

According to a report by self-media Z Potentials, LiblibAI's ARR has exceeded $1.5 million after just half a year of commercialization, ranking first in China.

LiblibAI also provides member users with a LoRA model training function, which allows users to train their own models by uploading images and setting parameters such as preset mode, bottom mold, number of single sheets, cycle rounds, and sample prompts. In May this year, LiblibAI upgraded the online image (WebUI) and added a variety of mainstream plug-ins. In June, LiblibAI updated the XL ControlNet plug-in model to greatly improve the quality of users' images and achieve accurate image control.

LiblibAI mainly encourages original model authors to continue to create, share high-quality models, and participate in operational behaviors such as interaction and promotion through the "Original Creator Incentive Program". The platform will issue subsidies to model creators on a monthly basis based on the comprehensive platform contributions of the model's user liking, user satisfaction, model weeping value and other dimensions.

In addition, LiblibAI has launched a series of hands-on classes and bootcamps for users to learn, and held creative design competitions, virtual photo contests, etc. to motivate authors to create.

In February 2024, LiblibAI passed the fourth batch of deep synthesis service algorithm filings by the Cyberspace Administration of China. In March of the same year, it became the first AI community in China to pass the national "Interim Measures for the Management of Generative Artificial Intelligence Services".

AI painting belongs to one of the branches of AIGC. Thanks to the extensive research and application of diffusion models, the current wave of AI painting enables text-generated image models to accurately grasp text information for creation. In August 2022, Stability AI released Stable Diffusion, an open-source image generation model that lowers the threshold for AI creation. Subsequently, AI painting quickly set off a spread on many online platforms in China, pushing AIGC creation to the common people. With the emergence of a large number of individual creators, platforms such as LiblibAI that bring together AIGC creators have also emerged.

However, the current business development trend of AIGC still faces challenges: it is still in the exploration stage in terms of commercialization, and the monetization method is relatively simple; AIGC's copyright and ethics issues have also been the focus of much controversy in the industry.

According to the iResearch report, the application of China's AIGC industry on the C-end is based on the content/social platform perspective, and the stickiness is formed through spontaneous communication between users in the form of a community, while the willingness and purchasing power of domestic users to pay for SaaS services are low.

Xia Ling, Partner of Mingshi Capital, said: "LiblibAI has become a leading community and tool platform in the field of Wensheng Graph in China in just one year since its establishment, relying on a deep insight into the pain points of professional users using Wensheng Graph models, as well as the rapid iteration of the team and the product itself. China has a strong supply capacity for content creative templates/fine-tuning models, and we believe that LiblibAI has the ability to provide a more complete and usable Wensheng graphic creation platform for global professional users compared to foreign platforms such as Midjourney and Adobe. Maxvision Capital will continue to support and help LiblibAI become a global leader in AI image generation. ”

Xiaohu Zhu, Managing Partner of GSR Ventures, said: "As the earliest investor in LiblibAI, GSR Ventures is delighted to have witnessed and accompanied the team's growth along the way. LiblibAI's reflection on the history of technological change and the deduction of the development path of AI technology reflect the company's profound thinking and decisive decision-making power in the strategic direction. In just one year, LiblibAI successfully topped the domestic list, delivering an excellent report card for the implementation of strategic decisions. We expect the company to continue to accelerate iteration in the future and become the best content production tool in the AI era. ”

Ma Xiaoyu, Managing Director of Gaorong Venture Capital, said: "LiblibAI has a huge passion for creating a platform that truly inspires and protects creators in the AI era, and has a keen insight into user needs, dares to think and do, and quickly iterates on community and tool products, which is very much in line with the portrait we were looking for a young team of AI Native at that time. We were impressed by LiblibAI's commitment to building and maintaining a better community atmosphere for model creators when the industry was in a hustle and bustle, whether it was the differentiated strategy of model supply in the early stage or the product roadmap for building more imaginative productivity tools for designers in the long term. We look forward to the company leveraging the power of the model ecosystem to provide more diversified aesthetic choices in the future, and empower more creators with new forms of AI interaction." ”

"There is still a lot of room for technological progress in generative AI, and AI images are the fastest productized, and this track will get the most lucrative compound interest growth in the future due to the double helix of technology and products," said Lulin Li, Vice President of Source Code Capital. In the main wave of rapid development of AI, the company's own ship speed is also fast enough - the starting point of the business accurately grasps the pain points of users and establishes an active model community and tool ecology, so that creators can quickly transform their own productivity and entertainment power with a steady stream of supply; The team is self-driven and diligent, especially when encountering the "road that no one has walked" that start-ups will always encounter, and they will also seek up and down and quickly get out of their own way. The company has always made the user experience the paramount criterion to create products that work better. The model community is only the starting point, and the end point is the infinite zero threshold for the expression of the "imagination" of users around the world. In the past year or so, the company has made every effort to achieve a high-quality start, and I believe that the company will be brave and sick in the future! ”