How Lovable Grew to a $500M Run Rate in 19 Months

Lovable had already won Product Hunt before the product called Lovable existed.

Its predecessor, GPT Engineer, had become one of the fastest-growing open-source coding projects on GitHub. Tens of thousands of developers starred the repository, about 27,000 people joined the waitlist for the commercial product, and the team had already launched successfully on Product Hunt.

But attention was not the same as product-market fit.

Founder Anton Osika later said the early commercial versions were good, not very good. People tried them, saw what was possible, and often failed to reach the moment where the product became useful enough to keep using.

Lovable spent more than a year turning that attention into a product people could actually finish something with.

The version that finally clicked launched in November 2024.

Four weeks later, Lovable said it had reached roughly $4 million in ARR. After about two months, it was at $10 million. By July 2025, TechCrunch reported $100 million ARR. By November, that had doubled to $200 million. In June 2026, Lovable said it had passed a $500 million annualized revenue run rate.

Two months later, Reuters reported that Lovable had raised $400 million at a $13.3 billion valuation.

The growth was extraordinary, but the useful part of the story is not the revenue chart.

Lovable combined six things unusually well:

• an audience built years before the breakout through open source
• product improvements that turned prototypes into usable apps
• strong retention once the product finally clicked
• a product whose output was naturally shareable
• founder, employee, creator, and community distribution
• integrations that improved the product and created new distribution partners

Market timing amplified all of it.

Lovable growth timeline

Date	Milestone
April 2023	Anton Osika publishes GPT Engineer on GitHub
June 2023	GPT Engineer passes 30,000 GitHub stars
January 2024	GPT Engineer wins Product Hunt
November 2024	Lovable launches publicly
December 2024	Lovable reaches roughly $4M ARR
January 2025	Lovable reaches roughly $10M ARR
February 2025	Lovable reaches roughly $17M ARR and 30,000 paying customers
July 2025	Lovable reaches $100M ARR
November 2025	Lovable reaches $200M ARR
February 2026	Lovable reports $400M ARR
June 2026	Lovable passes a $500M annualized revenue run rate
August 2026	Lovable raises $400M at a $13.3B valuation

Private-company revenue figures are mostly company-reported, so they should be treated as reported milestones rather than audited public financial results.

GPT Engineer created the audience before there was a business

Osika published the first version of GPT Engineer on GitHub in April 2023.

The idea was simple enough to spread quickly: give the model a prompt describing what you wanted to build, and it would generate a codebase.

It began as a weekend experiment.

Osika wanted to see how far large language models could go when asked to build software rather than answer coding questions one at a time.

The project exploded.

GPT Engineer passed 30,000 GitHub stars in June 2023 and later crossed 50,000. Osika's launch post gave developers an immediate demonstration of what the project could do without requiring them to read technical documentation first.

That open-source traction gave the future Lovable team four valuable assets:

• attention
• contributors
• product feedback
• an audience already interested in AI-assisted software creation

It still did not prove there was a business.

A developer can star an interesting repository and never use it again. A Product Hunt launch can produce a large spike of traffic without creating retention. A waitlist can show curiosity without showing willingness to pay.

Lovable still needed to discover what users wanted badly enough to come back for.

The first commercial ideas did not create lasting use

The team tested two reasonable product directions before Lovable found its eventual wedge.

The first was rapid prototyping.

Users could describe a product and generate an early version quickly.

The output looked impressive, but users often needed too much help to turn it into something complete. The product could create the first version of an idea without reliably getting the user all the way to a finished application.

The second idea focused more heavily on internal tools.

Businesses already paid developers to build dashboards, admin panels, and workflow applications. AI could clearly make that process faster.

It was a sensible market, but it was narrower than the team's larger ambition: let almost anyone create software.

Lovable used its waitlist to test different customer hypotheses instead of letting everybody into one generic beta. The team invited people who matched a hypothesis, watched what they tried to build, and asked what they had actually hoped to accomplish.

Osika has described this process in interviews with 20VC and Accel.

The important lesson was not that the first ideas were bad.

It was that early attention had hidden a weak activation and retention problem.

Why Lovable finally found product-market fit

The November 2024 launch reduced how much technical work users had to do themselves.

Lovable was no longer just trying to generate a convincing prototype. It was moving toward helping users build an actual product.

Its native Supabase integration was especially important.

Supabase gave Lovable users access to databases, authentication, storage, and backend functionality without requiring them to wire everything together manually.

Lovable later said that its Supabase integration coincided with growth from under $500,000 ARR to roughly $20 million ARR while the company was adding more than $1 million in ARR per week.

That does not prove Supabase caused the growth by itself. It does show how important the product shift was.

The value proposition had changed from:

AI can generate some code

to:

describe the software you want and get much closer to a working product

The product also kept improving rapidly after launch.

Stripe support arrived in December 2024. Visual Edits, which made it easier to adjust interfaces without repeatedly explaining every design change in a prompt, arrived in early 2025. GitHub syncing, previews, deployment, backend functionality, and other improvements steadily reduced the distance between an idea and a usable application.

The November launch mattered, but Lovable's growth came from the product continuing to improve immediately after it.

Retention separated Lovable from its earlier launches

Product Hunt attention was not new.

GPT Engineer had already performed well there.

The difference was that the Lovable product finally started retaining people.

In early 2025, Osika said Lovable had reached roughly 85% Day-30 retention while the company was around $17.5 million ARR.

That figure is founder-reported and unusually high, but it gets closer to the core of the story than launch-day traffic.

The first versions attracted curiosity.

Lovable increasingly created repeat behavior.

That distinction is important because many explanations of Lovable's growth focus on the channels that drove attention rather than the product quality that allowed that attention to compound.

The open-source audience, Product Hunt, social posts, and creators would have mattered far less if users still failed to reach a useful outcome.

Lovable arrived when demand was already exploding

Lovable also launched into unusually favorable market conditions.

ChatGPT had already taught millions of people that natural-language interfaces could replace complex workflows. GitHub Copilot had normalized AI-assisted programming. Cursor and other AI coding products were making developers dramatically faster.

At the same time, a much larger group of founders, designers, operators, marketers, and entrepreneurs wanted to build software without becoming professional developers.

Lovable packaged AI coding for that group.

Elena Verna, who later led product and marketing growth at Lovable, described the company's early challenge as capturing existing demand rather than creating a category from scratch. In Lenny's Podcast, she also warned against treating Lovable as a normal growth benchmark.

That caveat matters.

A founder can copy Lovable's experiments.

They cannot copy the exact moment when model quality, consumer awareness, and demand for AI-generated software all accelerated together.

How product-led virality helped Lovable grow

Lovable had another advantage: its output was inherently easy to show.

A user could:

• record a ten-second video of an app being built
• share a screenshot
• publish the application
• send someone a live link
• remix or demonstrate the result publicly

That made the product unusually easy to distribute on X, LinkedIn, YouTube, TikTok, Product Hunt, and inside communities.

A successful user did not need to write an essay explaining why Lovable was useful.

They could show the result.

That created a natural loop:

build something → share it → other people discover Lovable → they build something → they share it

Osika summarized the company's early acquisition strategy in an interview with Lenny Rachitsky: people loved the product, while the team created awareness by posting what it had shipped.

The phrase sounds simple because it leaves out the hard part.

You can only post what you shipped if you are shipping things people care about.

Shipping became a content strategy

Lovable did not begin with a conventional content-marketing machine.

Osika and other employees posted product changes themselves.

Small improvements could become demos. Larger launches became announcements. Integrations created another reason to post. A new editing mode could attract a new user, bring back an inactive one, and reassure an existing customer that the product was getting better.

The company did not need to invent topics disconnected from the product.

The roadmap created the content calendar.

That produced another loop:

ship → post → acquire users → get feedback → improve product → ship again

This was a particularly strong fit for an AI product because the category was changing quickly enough that people wanted to see what had improved that week.

Verna later described Lovable's growth work as roughly 95% innovation and 5% optimization, almost the reverse of some of her earlier growth roles.

The team was not only changing button colors or tuning landing-page conversion.

Growth people were helping build features, integrations, lifecycle experiences, and new ways for users to discover value.

Founder and employee distribution kept Lovable visible

Lovable's early social strategy was closely tied to shipping.

Osika posted product launches, growth milestones, lessons, technical breakthroughs, and examples of what people were building.

Employees also posted from their own accounts instead of routing every update through one polished company feed.

That created a more frequent and more credible stream of product information.

The content worked because it was usually attached to something real:

• a product improvement
• a customer creation
• a new integration
• a growth milestone
• a technical problem the team had solved

The strongest founder-led distribution is often just an information advantage.

The founder knows what changed inside the product before anybody else does.

Lovable turned that information into media.

Creators amplified a loop that already worked

Lovable was also unusually well suited to creator marketing.

The content format almost wrote itself:

1. describe an idea
2. show Lovable building it
3. reveal the finished app

Viewers could understand the product within seconds.

Verna has said short-form video worked particularly well because the novelty was immediately visible. She also said influencer marketing performed around ten times better than paid social, although Lovable has not published the spend or attribution methodology behind that claim.

The company tested widely instead of depending only on a few large creators.

Osika said Lovable had worked with more than 1,000 micro-creators and that fewer than 10% generated roughly 90% of the reach.

Those numbers are founder-reported, but the underlying strategy is clear.

Lovable treated creators like a portfolio.

Most would produce modest results. A small number would produce disproportionate reach.

The team also personally onboarded some early creators so they understood the product well enough to make something useful with it.

The best creator content looked like a genuine build, not a person reading a sponsorship script.

Community credits turned organizers into onboarding partners

Lovable gave product credits to hackathons, meetups, and community organizers.

That did more than generate awareness.

It removed the payment decision before the user had experienced the product.

A meetup organizer could introduce Lovable, help participants through failed prompts, answer basic questions, and encourage everybody to finish something during the session.

Lovable supplied the product and credits.

The organizer supplied the audience, context, and hands-on onboarding.

Verna has argued that those credits should effectively be viewed as marketing spend.

The model created another loop:

free credits → group builds together → more users reach activation → finished projects get shared → more people discover Lovable

By November 2025, Lovable said its community had reached 100,000 members and was running between 10 and 30 meetups or hackathons each week in its one-year retrospective.

The March 2026 SheBuilds program expanded the model further, with organizers running more than 120 gatherings across over 40 countries.

There is no public evidence that one of these programs caused a specific revenue increase.

Their importance is structural: Lovable found a way to turn product access into community distribution and assisted onboarding at the same time.

Integrations doubled as distribution

Lovable's integrations improved the product, but some also created distribution.

Supabase is the clearest example.

Lovable users needed databases, authentication, and backend infrastructure. Supabase benefited when more Lovable users built real applications.

That meant both companies had an incentive to explain the integration and show users what they could build together.

Lovable's own retrospective on reaching $10 million ARR also points to co-marketing with companies such as Supabase, Replicate, and Resend.

The loop looked like this:

Lovable integrates another product → the combined workflow becomes more useful → both companies have something to promote → each company exposes the other to its audience

For early-stage software companies, integrations are often treated only as product features.

Lovable also used them as distribution surfaces.

Product Hunt helped, but it did not create the business

Lovable launched on Product Hunt in November 2024 and finished as Product of the Day with nearly 1,500 votes and hundreds of comments.

That certainly concentrated attention.

But Product Hunt was not the explanation for the growth.

GPT Engineer had already performed well on Product Hunt before Lovable found strong retention.

The Hacker News launch thread for Lovable received only 32 points and 13 comments.

Osika later said the company could have received far more press around the launch.

That is useful context.

Lovable did not suddenly become huge because one launch channel exposed it to the world.

It already had years of open-source attention, a large waitlist, an audience interested in AI coding, and a product that had finally become much easier to finish work with.

Product Hunt helped focus that demand.

It did not create it.

Lovable priced for adoption, then usage

AI app building has real model and infrastructure costs.

A free plan can create a huge acquisition funnel and a huge bill at the same time.

Lovable used credits to control that exposure.

Free users could experience the product. More intensive builders moved into paid subscriptions or bought more usage.

The company also made decisions that favored collaboration and adoption over maximizing seat revenue.

In June 2025, Lovable removed a higher-priced Teams tier and moved collaboration features into Pro. Osika said the decision reduced ARR by $1.5 million in a single day.

That is his own account, not an audited result.

Later pricing changes included annual plans, credit rollovers, top-ups, and removing per-seat pricing.

Verna told Stripe that repeat top-up purchases eventually became comparable to or better than subscription renewals.

The model increasingly tied expansion to how much value people got from building rather than simply how many seats a company purchased.

Templates, public projects, and SEO became additional loops later

Lovable eventually added templates, public projects, remixing, and branding on free published sites.

Each gave the company another potential acquisition surface.

A public project could be discovered.

A template could rank in search.

A free site could expose the Lovable brand.

A project could be remixed into a new project.

These features matter, but their role in the origin story is easy to exaggerate.

Lovable has not publicly disclosed how many early signups came from templates, remixing, badges, or SEO.

Verna has also described SEO as a useful baseline rather than the reason a company wins.

The first breakout is better explained by product quality, existing demand, word of mouth, social distribution, creators, community, and integrations.

The formal growth team came after product-market fit

Verna joined Lovable in May 2025.

She recalled joining when the company had around 20 employees and roughly $40 million in ARR.

By later in 2025, Lovable was hiring growth product managers, engineers, and data scientists and building a more formal system across creators, integrations, events, lifecycle, partnerships, and enterprise sales.

That organization should not be projected backward.

Lovable did not need a large growth department to find its first loop.

The formal growth machine came after the product had already demonstrated unusually strong demand.

The company hired more heavily once global scale, infrastructure, security, enterprise sales, and lifecycle optimization became larger problems.

The five growth loops behind Lovable

Lovable's growth is easier to understand as a set of reinforcing loops than as a list of channels.

1. Open-source loop

GPT Engineer → GitHub attention → contributors and audience → commercial waitlist → Lovable users

Open source created distribution before the company had found the final product.

2. Product loop

better reliability → more completed apps → stronger retention → more revenue → more product investment

This is the loop that the first commercial versions were missing.

3. User-generated-content loop

user builds app → shares result → viewer discovers Lovable → viewer builds app → shares result

Lovable's output acted as proof of the product.

4. Shipping loop

team ships → founder or employee posts demo → users discover or return → team gets feedback → team ships again

The roadmap produced the marketing material.

5. Ecosystem loop

Lovable integrates Supabase, Resend, Stripe, and others → combined workflow improves → partners promote use cases → more Lovable adoption

Integrations created both product value and distribution.

These loops reinforced one another.

That is why reducing Lovable's growth to Product Hunt, X, creators, or one viral launch misses most of the story.

The revenue numbers are large and mostly company-reported

Lovable said it reached roughly $4 million ARR four weeks after its November 2024 launch and $10 million after about two months.

TechCrunch reported roughly $17 million ARR and 30,000 paying customers in February 2025.

In July 2025, TechCrunch reported $100 million ARR.

By November, that figure had reached $200 million.

Lovable reported $400 million ARR in February 2026.

In June 2026, the company said it had passed a $500 million annualized revenue run rate and was generating around one million new projects every week.

Reuters reported in August 2026 that revenue had nearly tripled from the $200 million level and was tracking toward $600 million.

That is not the same thing as confirming $600 million ARR.

The financing milestones are easier to verify.

Lovable raised $200 million at a $1.8 billion valuation in July 2025, $330 million at a $6.6 billion valuation in December, and $400 million at a $13.3 billion valuation in August 2026.

Private-company revenue numbers are not audited public filings.

They are still useful signals when described using the same wording and dates as the original sources.

Lovable still has expensive product problems

Lovable's growth does not mean every part of the product is solved.

Reddit and Product Hunt discussions repeatedly mention credit burn, failed repair loops, and projects becoming more difficult to manage as they grow.

Security has also become part of the story.

Researchers and publications have reported vulnerabilities and disputed data-exposure incidents. Lovable has published responses and added additional security features.

That makes blanket claims such as "Lovable makes production-ready software automatically" hard to defend.

Its clearest strength is reducing the effort required to create the first working version of software.

Maintaining a large production system is a different problem.

That distinction will matter more as Lovable moves deeper into business-critical applications and enterprise customers.

What other startups can actually learn from Lovable

The most useful part of Lovable's story is not the $500 million figure.

It is what happened before the chart became impressive.

The company had an audience and still failed to create lasting use.

It launched successfully and still kept changing the product.

It used a waitlist to test specific customer hypotheses instead of confusing attention with product-market fit.

Once retention improved, its distribution advantages started compounding.

The strongest lessons are practical:

• build an audience before you desperately need distribution
• do not confuse launch attention with retention
• make the product output easy to show
• turn product shipping into content whenever the changes are genuinely useful
• use creators as a portfolio rather than expecting every partnership to work
• give communities a reason to help users reach activation
• choose integrations that can create both product value and shared distribution
• keep improving the product after the launch appears successful from the outside

Lovable is still a poor normal-growth benchmark.

It caught a rare market wave at exactly the right time and entered that moment with years of open-source attention behind it.

Other companies cannot copy that timing.

They can copy the way Lovable turned product improvements, user output, employees, creators, communities, and partners into reinforcing distribution loops.

Frequently asked questions

How did Lovable grow so fast?

Lovable combined an audience built through the open-source GPT Engineer project with a much stronger full-stack product, unusually strong market timing, founder-led social distribution, rapid product releases, creator marketing, community events, and integrations such as Supabase. Once retention improved, those channels began compounding rather than producing one-time traffic spikes.

When did Lovable launch?

Lovable publicly launched in November 2024. The company grew out of GPT Engineer, the open-source AI coding project Anton Osika first published in April 2023.

How fast did Lovable reach $10M ARR?

Lovable said it reached roughly $10 million ARR about two months after its November 2024 launch.

How did GPT Engineer help Lovable grow?

GPT Engineer gave the team an audience before Lovable existed. The open-source project attracted tens of thousands of GitHub stars, contributors, feedback, and a large waitlist of people already interested in building software with AI.

Did Product Hunt drive Lovable's growth?

Product Hunt helped concentrate attention, but it did not create Lovable's product-market fit. GPT Engineer had already performed well on Product Hunt before the team found strong retention. The stronger product, existing audience, word of mouth, social distribution, creators, and market timing mattered more.

How did Lovable use influencer marketing?

Lovable worked with a large number of micro-creators rather than relying only on a few large accounts. Anton Osika said the company had worked with more than 1,000 micro-creators and that fewer than 10% generated roughly 90% of the reach. Elena Verna later said influencer marketing performed substantially better than paid social, though Lovable has not publicly released the underlying attribution methodology.

What role did Supabase play in Lovable's growth?

Supabase helped Lovable move beyond frontend prototypes by giving users access to databases, authentication, storage, and backend functionality. Lovable later said its Supabase integration coincided with growth from under $500,000 ARR to roughly $20 million ARR.

Did SEO drive Lovable's early growth?

There is no public attribution showing that SEO drove Lovable's initial breakout. Templates, public projects, and search became additional acquisition surfaces later. Founder and executive accounts emphasize product quality, word of mouth, social posts, creators, community, and integrations more heavily.

What is Lovable's latest reported revenue?

Lovable said it had passed a $500 million annualized revenue run rate in June 2026. Reuters reported in August 2026 that revenue was tracking toward $600 million, but that is not the same as confirming $600 million ARR.

How much has Lovable raised?

Lovable raised $200 million at a $1.8 billion valuation in July 2025, $330 million at a $6.6 billion valuation in December 2025, and $400 million at a $13.3 billion valuation in August 2026.

Sources

• GPT Engineer, GitHub
• Anton Osika launches GPT Engineer, LinkedIn
• Anton Osika on Lovable's early product tests, 20VC
• Anton Osika on Lovable's product and growth, Lenny's Podcast
• Lovable's November 2024 launch, LinkedIn
• Lovable launch discussion, Hacker News
• Lovable's first year, Lovable
• Lovable on its Supabase integration, Lovable
• Stripe customer story with Elena Verna, Stripe
• Lovable reaches roughly $17M ARR, TechCrunch
• Lovable crosses $100M ARR, TechCrunch
• Lovable reaches a reported $500M annualized run rate, TechCrunch
• Lovable raises at a $13.3B valuation, Reuters
• Lovable customer discussions, Reddit