Lovable pricing

About

Lovable is an AI full-stack app builder: you describe an app in natural language and it generates a working React + Supabase application — auth, database, hosting and connectors included — that you can export to GitHub and own outright. It positions itself less as a coding copilot for engineers (its FAQ explicitly contrasts itself with Cursor) and more as a “vibe coding” tool that lets entire teams — PMs, designers, marketers, operators — build real software while engineers keep control via standard code and GitHub PRs.

Commercially, Lovable runs a freemium subscription metered by credits. Every AI message spends a credit; the Free tier hands out daily credits, while paid tiers (Pro and Business) bundle a pool of 100 monthly credits plus daily credits, with on-demand top-ups and credit rollovers when teams need more. Notably, paid plans are not per-seat — a workspace’s credits are shared across unlimited users, so customers pay for AI capacity rather than headcount. The ladder runs Free ($0) → Pro ($25/mo) → Business ($50/mo) → Enterprise (custom platform fee), with student and K12-classroom programs on the side.

For the most current information on Lovable’s pricing and market position, visit Lovable.

Pricing summary : How Lovable’s pricing model works

Lovable’s bill is built from two dimensions: a flat monthly subscription tier and credit consumption (each AI message costs a credit). The subscription sets the included credit pool and feature gate; usage above the pool is handled by on-demand credit top-ups, while unused credits roll over on paid plans. Unusually for the category, the subscription is not priced per seat — a workspace’s credits are shared across unlimited users, so adding teammates doesn’t add cost; running out of credits does.

• Free — $0/mo, no card. 5 daily credits (up to 30/month).
• Pro — $25/mo, or $21/mo billed annually (save $50/yr). 100 monthly credits + 5 daily credits (up to 150/month), credit rollovers, on-demand top-ups.
• Business — $50/mo, or $42/mo billed annually (save $100/yr). 100 monthly credits + Pro features, plus SSO, team workspace, role-based access and a security center.
• Enterprise — custom “platform fee” based on company size (covers all employees), with volume-based credit pricing, SCIM and audit logs.
• Student / classroom — verified students get up to 50% off Pro (~$12.50/month for one year); K12 grades 6-12 get free classroom access via imagi Edu.

What makes this different: the seat decoupling. Credits — not users — are the value metric, so the same $25 or $50 plan covers a solo builder or a whole team, and the lever that moves the bill is how many AI messages the workspace sends, not how many people it has.

Pricing by product

Lovable (subscription plans)

Education programs

Sales motions across products: PLG / self-serve for Free, Pro and Business (sign up and pay online); sales-led for Enterprise (book a demo, custom platform fee) and Lovable for campus.

Hidden costs : What Lovable users actually pay

The sticker price ($25 or $50/mo) is the floor, not the bill. Lovable meters work in credits, and since July 2025 a credit is no longer a flat “one message” — it is complexity-weighted. A simple chat reply is ~1 credit, a minor UI tweak ~0.5, and a complex build step (“add authentication with sign-up and login”) can cost 1.5+ credits. A from-scratch app skeleton commonly burns ~2 credits before you have changed anything. That makes the 100 monthly credits on Pro/Business easy to exhaust during active iteration — and vague prompts burn faster because they force more rounds.

When the pool runs out, the lever is on-demand top-ups, which are priced differently by tier: Pro buys roughly $15 per 50 credits, Business roughly $30 per 50 credits, with purchased credits expiring after 12 months. Separately, apps you ship can run on Lovable Cloud (managed Supabase-style backend), which carries its own usage-based bill for storage, auth and serverless functions beyond a free allowance — a line item that lives outside the subscription entirely.

The combination of a flat subscription, a variable complexity-weighted credit meter, and a separate usage-based Cloud bill is exactly what reviewers flag as making month-to-month spend hard to forecast. Credit rollovers on paid plans soften this for bursty teams, but the burn rate is set by prompt quality and app complexity, not headcount.

Want to estimate your own Lovable bill? Use the Lovable pricing calculator to model credit burn and top-ups against your usage patterns.

Pricing evolution : Lovable pricing history and changes

Cadence

Tracked range: 2024 Q4–present. The main lovable.dev/pricing surface is not directly archived (Wayback snapshots redirect to the homepage); dated changes are corroborated by Lovable’s blog and secondary coverage.

Notable changes

• 2024-11-21 — Lovable introduces subscription tiers (Free / Starter $20 / Launch $50 / Scale $200) priced by successful edits/messages per day or month, with early-access users grandfathered (Lovable blog).
• 2025-07-24 — Pricing shifts from flat “1 credit = 1 message” to complexity-weighted credit consumption (chat ~1, tweaks ~0.5, complex builds 1.5+; error-fix retries generally free) as Lovable goes agentic and crosses $100M ARR — billed as the fastest software company ever to that mark (Tech.eu, Sacra).
• 2025–2026 — Tier names converge on the current Free / Pro $25 / Business $50 / Enterprise ladder, with the defining move being flat per-workspace pricing shared across unlimited users rather than per-seat licensing.

What’s unique : Lovable’s distinctive pricing mechanics

1. Seats are decoupled from price — you buy AI capacity, not headcount. The single most distinctive choice is that Pro ($25) and Business ($50) are per-workspace, not per-user: a workspace’s 100 monthly credits are shared across unlimited collaborators. In a category where Cursor, GitHub Copilot and most dev tools charge per seat, Lovable charges for the thing it actually spends money on — model inference — and lets the whole team (PMs, designers, marketers) onto the same plan for free. The value metric is the credit, not the login.

2. Complexity-weighted credits, not flat per-message billing. Since July 2025 a credit is a unit of work, not a unit of interaction: a one-line tweak can cost ~0.5 credits, a heavy build step 1.5+, and failed-build “try to fix” retries are generally free. This aligns spend with the model cost of the request and softens the bad-UX failure mode of charging users for the AI’s own mistakes — a deliberate contrast to flat-message competitors.

3. A daily-credit floor plus monthly pool, rollovers and tier-priced top-ups. Every plan layers a refilling daily allowance (5/day) on top of the monthly pool, so even Free users get a steady trickle and paid users have a cap of up to 150/month from dailies alone. Unused monthly credits roll over, and overflow is handled by on-demand top-ups priced by tier (~$15/50 on Pro, ~$30/50 on Business) rather than a forced upgrade — a continuous-meter design wrapped around a subscription shell.

Strengths & weaknesses

Billing UX : Lovable billing controls and transparency

• Monthly / Annual toggle — every paid card flips between monthly and annual billing inline, showing the discounted per-month rate ($21 Pro, $42 Business) and an explicit “Save $50 / Save $100” annual savings label.
• Credit pools + daily credits — the unit of usage is a credit (one per AI message); plans expose both a monthly pool (100 on paid) and a daily allowance (5/day, capped per tier) so users can see exactly what they have.
• Credit rollovers — unused monthly credits carry forward on paid plans, smoothing bursty usage.
• On-demand credit top-ups — when a workspace exhausts its pool, it can buy more credits without changing tiers, avoiding a hard stop.
• Shared-workspace billing — credits are pooled across unlimited users, so adding collaborators is a no-cost action; the billing surface tracks capacity, not seats.
• Self-serve student verification — students unlock the ~50% discount by verifying an educational email or student ID directly, no sales contact.
• Enterprise governance controls — SSO/SAML, SCIM provisioning, role-based access, audit logs and publishing/sharing controls gate the Business and Enterprise billing tiers.

Strategic wins : Why Lovable’s pricing decisions worked

1. Picking credits over seats unlocked viral team adoption

By making the value metric a credit and the plan per-workspace, Lovable let entire teams pile onto one $25 plan for free. That removed the classic per-seat tax that slows bottoms-up adoption inside companies, and it matched the cost driver (model inference) to the price. Combined with the open-source GPT Engineer halo, it is a textbook case of the shift away from per-user licenses — and it helped Lovable reach a reported $100M ARR roughly eight months after launch, billed as the fastest software company ever to that mark.

2. Complexity-weighting the credit aligned price with cost without bill-shocking on failures

The July 2025 move from flat per-message to complexity-weighted credits — while keeping error-fix retries free — tightened the link between what a request costs Lovable and what the user pays, without punishing people for the AI’s own mistakes. It’s a pragmatic middle ground between flat subscriptions and pure metering, and a useful data point for anyone weighing outcome- and usage-based pricing for AI products.

3. The free daily-credit trickle keeps the funnel full

A no-card Free tier with 5 refilling daily credits gives curious non-developers enough rope to ship something small and feel the magic, then hit a natural ceiling that nudges them to Pro. Choosing a metric that’s both the demo currency and the billing currency is a clean example of picking the right usage metric: the same credit that sells the product is the one that monetizes it.

Areas to improve : Gaps in Lovable’s pricing approach

1. Credit burn is genuinely hard to predict

The most consistent reviewer complaint is forecasting: complexity-weighted credits mean an identical-looking task can cost 0.5 or 1.5+ credits, and vague prompts multiply the rounds. With only 100 monthly credits on Pro/Business, heavy iteration can drain the pool well before month-end. Better in-product burn estimates and a clearer “credits remaining at this rate” signal would directly reduce the cost-unpredictability bill shock the model invites.

2. The Lovable Cloud bill hides outside the subscription

Apps that ship on Lovable Cloud accrue separate usage charges (storage, auth, serverless functions) that aren’t reflected on the pricing page’s tier comparison. For a product sold to non-developers, surfacing this second meter earlier — ideally inside the same credit dashboard — would prevent surprise downstream costs.

3. Opaque Enterprise and an inverted top-up curve

Enterprise is quoted as a “platform fee based on company size” with no public anchor, which makes it hard for mid-market buyers to self-qualify. And top-ups cost roughly 2× more per credit on Business ($30/50) than on Pro ($15/50) — counterintuitive for the higher-priced tier — which deserves either a published rationale or a flatter curve.

Key takeaways

1. Pay for capacity, not seats. Pro ($25) and Business ($50) are per-workspace with credits shared across unlimited users — adding teammates is free, running out of credits is what costs money.
2. A credit is a unit of work, not a message. Since July 2025 credits are complexity-weighted (chat ~1, tweaks ~0.5, complex builds 1.5+), and failed-build error fixes are generally free.
3. The sticker price is a floor. 100 monthly credits go fast under iteration; expect tier-priced top-ups (~$15/50 on Pro, ~$30/50 on Business) and a separate Lovable Cloud usage bill.
4. Free is a real funnel, not a trial. Every plan including Free gets refilling daily credits (5/day), giving non-developers enough to ship something before hitting a natural upgrade ceiling.
5. Growth followed the metric. Choosing credits-over-seats helped Lovable reach a reported $100M ARR in ~8 months — a marquee proof point for usage-aligned pricing in AI dev tools.

UBP implications

1. Decoupling seats from price can be a growth weapon, not just a billing detail. When your real cost is model inference, per-seat pricing taxes the wrong thing. Lovable’s per-workspace prepaid-credit pool let whole teams adopt for free and made the credit the single metric that both demos and monetizes the product.
2. Complexity-weighted metering is the honest middle path — but it shifts the forecasting burden to the buyer. Charging more for harder work (and nothing for the AI’s own retries) aligns price with cost, yet it makes spend non-linear. If you adopt it, invest equally in burn-rate visibility, or you import the bill-shock problem you were trying to avoid.
3. A subscription shell around a continuous meter needs a complete meter. Lovable’s hidden third bill (Cloud usage) shows the trap: a clean credit story is undermined when material usage costs sit outside the dashboard. Surface every meter the customer can trigger, or predictability evaporates.

Sources

• Lovable pricing page (accessed 2026-06-08)
• Lovable plans and credits (docs) (accessed 2026-06-08)
• Lovable for enterprises (accessed 2026-06-08)
• Lovable for students (accessed 2026-06-08)
• Lovable for the next generation (classroom / imagi Edu) (accessed 2026-06-08)
• Lovable blog: targeted edits & pricing updates (accessed 2026-06-08)
• Lovable changelog (accessed 2026-06-08)

Bottom line

Lovable’s pricing is the clearest production-scale bet on credits-over-seats in AI dev tools: one flat $25 or $50 workspace plan, shared across an entire team, metered by complexity-weighted credits rather than logins. That choice — pay for AI capacity, not headcount — helped it reach a reported $100M ARR in roughly eight months. The catch is forecasting: between a non-linear credit meter, tier-priced top-ups, and a separate Lovable Cloud bill, month-to-month spend is genuinely hard to predict, which is exactly the trade-off usage-aligned pricing demands you manage in the open.

Want to estimate your own bill? Try the Lovable pricing calculator, or compare Lovable against other developer-tools companies in the pricing blueprint.

Pricing timeline : Major events on a vertical axis

Each milestone below corresponds to a public pricing change, product launch, or material adjustment. Major events use a filled marker; minor adjustments use a faded one.

Current pricing captured

Jun 2026

Free $0 (5 daily credits, up to 30/mo); Pro $25/mo ($21 annual); Business $50/mo ($42 annual); both paid tiers 100 monthly credits + daily credits, shared across unlimited users. Enterprise custom platform fee by company size. Students up to 50% off Pro (~$12.50/mo). Note: lovable.dev/pricing snapshots in the Wayback Machine redirect to the homepage, so the main pricing surface was not directly archived; dated changes above are corroborated by Lovable's blog and secondary coverage.

Shift to granular credit consumption + agentic mode

Jul 2025

Lovable moved from 1 credit = 1 message to complexity-weighted credit consumption (simple chat ~1 credit, minor tweaks ~0.5, complex builds 1.5+; failed-build error fixes generally free) alongside its agentic relaunch. This coincided with crossing $100M ARR ~8 months after launch. Sources: https://lovable.dev/blog/agent ; https://sacra.com/c/lovable/

Lovable launches paid tiers (from GPT Engineer)

Nov 2024

GPT Engineer relaunched as the Lovable web app and introduced subscription tiers: Free ($0, 5 messages/day), Starter ($20/mo), Launch ($50/mo) and Scale ($200/mo), metered by successful edits/messages per day or month. Early-access users grandfathered. Source: https://lovable.dev/blog/2025-01-13-targeted-edits-supabase-integration-improvements-and-pricing-updates