ARR exceeded $100 million in eight months, and Lovable secured $200 million in Series A financing.

Since the product was launched in November 2024, in just eight months, AI programming startup Lovable has secured two rounds of financing. In February 2025, it received $15 million in Pre-A round financing, led by top Swedish venture capital firm Creandum, with participation from Adam D’Angelo and Thomas Wolf (co-founder of Hugging Face), among others.

In July, it secured $200 million in Series A financing led by Accel. 20VC, ByFounders, Creandum, Hummingbird, and Visionaries Club also participated in this round. After this round of financing, Lovable has raised approximately $224 million in total, and its valuation has reached $1.8 billion.

Angel investors who participated in the Series A financing also include Mati Staniszewski of ElevenLabs, Sebastian Siemiatkowski of Klarna, Victor Riparbelli and Steffen Tjerrild of Synthesia, and Nik Storonsky of Revolut.

Source: Lovable

The driving force behind Lovable's rapid acquisition of large - scale financing is its extraordinary growth rate. When it received the Pre - A round of financing in February, it had 500,000 users, 30,000 paying users, and an ARR of $17 million. After six months, its user base reached 2.3 million, with over 180,000 paying users, and its ARR exceeded $100 million. Its achievement of exceeding $100 million in ARR in eight months has outpaced Cursor (12 months), Wiz (18 months), and Deel (20 months), making it the record - holder for this revenue growth rate.

Source: AI Product Ranking

According to the statistics of the media's AI Product Ranking, Lovable's monthly visits reached 35.88 million in June 2025, while the benchmark product Cursor had 18.87 million visits. Anysphere (Cursor's parent company) received a $900 million round of financing in June, with a valuation as high as $9.9 billion.

01

Assembled the founding team in 2 hours, using Vibe Coding to help 99% of non - professional people build applications with low barriers

AI coding is not a hot field that emerged in 2025. In 2024, companies such as Poolside, Magic, Cognition, Augment, Anysphere, and Windsurf (just acquired) had already received single - round financing of over $100 million. Anysphere's latest $900 million financing has clearly established a leader in this field.

Model provider Anthropic has also launched Claude Code. Since its launch in May, its weekly downloads have increased six - fold, contributing over $200 million in ARR.

Why can Lovable find a way out in this crowded field? Most other AI coding companies focus on the 1% of professional programmers, while Lovable targets the other 99% of non - professional people, occupying the Vibe Coding track.

So - called Vibe Coding means allowing people to use natural language to give instructions to AI tools, which then generate, optimize, and debug code to quickly build prototypes of websites and other applications. On the one hand, it enables people without programming knowledge to try out their ideas. On the other hand, it allows programmers to quickly iterate on their application prototypes, improving development efficiency.

Lovable's origin is also quite interesting. Its founder, Anton Osika, was formerly the CTO of Depict. After three years, he felt that e - commerce optimization was not the career he wanted to devote his life to. At that time, ChatGPT emerged, and he began to explore what kind of complex systems could be built with AI.

One weekend, he created a simple prototype: "Generate a usable application with plain language." He named this prototype GPT Engineer, posted it on GitHub, and then went to sleep.

Within a few weeks, this project attracted hundreds of thousands of users, and its GitHub stars exceeded 50,000. His inbox was flooded, and most of the senders were not professional programmers but people who had never written a single line of code but had built their first applications with it.

Anton Osika realized that he had found a new market: helping the 99% of non - programmers to realize their great ideas and solve problems that had not been solved before. These people include not only developers but also entrepreneurs, designers, business owners, product managers, and operators.

The mission of AI should not be to help engineers code faster but to help everyone else create.

At 6 a.m., he called co - founder Fabian Hedin. As they walked and talked, Lovable was born by 8 a.m.

Source: Lovable

Anton Osika previously worked at the European Organization for Nuclear Research (CERN) and Sana Labs. His co - founder, Fabian Hedin, is a super programmer who previously developed the interactive interface for the computer used by famous physicist Stephen Hawking and collaborated with former SpaceX engineers on wheelchair technology.

The paradigm shift brought by Lovable lies in creating new markets. For example, before Canva emerged, there were 5 million 'designers' using Photoshop. After Canva lowered the design threshold, it gave rise to 220 million designers, expanding the market 44 times.

Source: Lovable

There are 47 million software developers globally, while Lovable's potential users could reach 1 billion. In the past eight months, many people have used Lovable to build their software businesses. For example, Caio Moretti's product built with Lovable generated $3 million in revenue within 48 hours. Airweave (an intelligent agent search tool for internal databases) was successfully selected by Y Combinator.

Lovable doesn't help programmers write code but helps 'amateurs' achieve one - stop application development

AI coding tools like Cursor are aimed at programmers, helping them generate or modify code more efficiently, and these tools integrate traditional coding tools.

Lovable's goal is to achieve one - stop application development. It integrates with database Supabase and code repository GitHub to help non - programmers build applications from scratch, ultimately resulting in a usable application (whether it is sophisticated or rudimentary).

Source: Lovable

Users only need to enter an idea in the chat box on the website to start their creative journey. The platform also supports uploading various files such as pictures, documents, and flowcharts. Users with some technical background can also import their works from the cloud - based UI/UX design and prototyping tool Figma, providing a framework for the AI tool to build the logic behind the application.

After entering the Vibe Coding field, Lovable's 2.0 product has expanded its user base to enterprise teams

Simultaneously with the announcement of the new financing, Lovable launched its 2.0 version. In this version, users can have more precise control over the quality of applications. It seems that its user base has expanded from 'amateurs' to small teams within companies, encroaching on Cursor's territory.

For example, Lovable has introduced 'workspaces' that allow multiple people to collaborate on projects. The Pro version allows 2 - person collaboration, while the Teams version allows 20 - person collaboration, clearly targeting small enterprise teams.

There is also the intelligent agent mode, which can reduce the error rate of application building by 90%. Specifically, it can search code libraries, read files on demand, check logs and network activities, and conduct real - time network searches.

Of course, there is also visual editing, which allows users to edit any style visually, enabling faster and more precise modifications. It is obviously targeted at non - programmer users.

Source: Lovable

Then there is the developer mode and security scanning. The developer mode allows users to directly edit project code, while security scanning checks for security vulnerabilities in the application. The former is basically for programmers, and the latter ensures the quality of the application when it goes live.

From the functional changes in the 2.0 version, we can see that Lovable's strategy is to, on the one hand, continue to improve the Vibe Coding experience and ensure the quality of applications developed through Vibe Coding. On the other hand, it is encroaching on the territory of other AI programming applications, such as the intelligent agent mode (Devin's territory) and the editable developer mode (Cursor's territory). This will allow it to maintain differentiated competition while strengthening its competitive advantage.

AI tools are driving the emergence of a large number of 'nano unicorns'

Silicon Valley investment firm a16z has summarized a type of startup called 'nano unicorns'. They have lean teams, high revenues, and rapid growth. For example, Anysphere (Cursor's parent company) achieved $300 million in ARR with only 20 employees, and currently, Lovable has no more than 50 employees.

These companies use AI to enhance team capabilities in all functions related to engineering, product development, and customer - facing operations. The same goes for Cognition, where each employee'manages' 5 - 20 Devin intelligent agents, using 'intelligent agent employees' to increase the company's 'combat power'. In this model, AI is not an auxiliary tool but is regarded as infrastructure and the core of work.

AI programming tools can be used by people to validate ideas and iterate on product prototypes, and they are also powerful tools to significantly improve the efficiency of technical teams. In the future, there may be more 'nano unicorns' composed of small elite teams.

02

The rapid growth of AI applications is driven by innovation

In 2021, Deel and Wiz broke the record for reaching $100 million in ARR in 20 months and 18 months respectively. From 2024 to 2025, Cursor and Lovable have pushed the revenue growth rate to an astonishing level in 12 months and 8 months respectively.

Why is the revenue growth rate of AI applications so 'absurdly' fast? The growth rate in 2021 was driven by demand under the special circumstances at that time. The speed in 2025 is driven by innovation. The capabilities of AI have transformed the appearance of applications and entered many new markets.

So why has AI coding created such an explosion in the field of AI applications? Because it has found value points. In fact, as long as value can be created, people are willing to pay in the AI era, just like in the fields of AI law and AI search. For example, Harvey AI has an ARR of over $50 million and a valuation of $3 billion.

Although we have seen giant startups like OpenAI and Anthropic, and tech giants like Google and Alibaba are also very active, innovation is a game where the fast fish eat the slow fish, and startups composed of a small number of elites are faster. Moreover, AI technology is still in its early stages, and as long as there is innovation, there will always be opportunities, especially in the application layer. In 2024, the number of AI application - layer companies receiving over $100 million in financing exceeded the total of all other fields, which has proven this point.

The opportunities in the application layer lie in vertical fields. As long as one can create an experience 10 times better than others in a vertical field or capture a new group of users, bringing real productivity improvements, there is a great chance of success.

Innovation has always been the strength of Chinese product managers. This has been repeatedly proven. Now, on the eve of the great explosion of AI applications, the opportunities should not be wasted.

This article is from the WeChat public account "Alpha Commune" (ID: alphastartups). Author: Alpha Commune, which discovers extraordinary entrepreneurs. Republished by 36Kr with authorization.