Bolt vs Lovable vs v0 for Client MVPs
Clients ask for a “quick landing page / waitlist / internal tool.” You can burn a weekend in Cursor — or ship a demoable web app in an afternoon if you understand credits, export, and what “MVP” actually means in the SOW.
This is a job-based pick between three prompt-to-app lanes: Bolt.new, Lovable, and v0. Not a hype demo reel. No invented “shipped 40 client apps” stories.
Affiliate disclosure: builder links may become affiliate/referral later. Plan numbers below are from vendor pricing pages as of 21 Sep 2026 (SAST) — re-check before you buy.What “client MVP” means here — and what it is not
Here, MVP means a demoable web surface the client can click: marketing site, waitlist + email capture, simple internal CRUD, or a UI shell that proves the concept. It is not a bank-grade production system, not a full design system, and not a substitute for a senior engineer when the brief needs auth hardening, complex payments, or regulated data.
Good freelancers sell the demo honestly: “This proves the flow; production hardening is a separate phase.” Bad freelancers let credits eat the fee while chasing pixel-perfect forever-apps inside a free tier.
If the client only needs a deck, use a deck tool — see Gamma AI for client decks. If they’re still learning how to sell AI services at all, start with AI freelance for beginners.
Bolt.new — browser build, token burn, Pro ladder
Bolt.new pricing (checked 21 Sep 2026):- Free: 300K tokens daily limit, 1M tokens per month, Bolt branding on sites, 10MB upload limit, hosting included with usage caps listed on the page.
- Pro: $25/month (billed monthly) — no daily token limit, starts at 10M tokens/month, unused tokens roll over (with paid-subscription rules), custom domain, no Bolt branding, higher upload/hosting limits, AI image editing, and more as listed.
- Teams / Enterprise: higher seat and governance options on the same page ($30/member framing for Teams; custom Enterprise).
Bolt lives in the browser. Token burn scales with project size because the tool syncs file context to the model — big projects get expensive per message. For freelancers: keep demos small, reset or branch cleanly, and don’t iterate forever on Free when Pro’s monthly allotment is what the SOW needs.
Lovable — chat-to-app, shared credits, ownership notes
Lovable pricing is credit-based, not “unlimited magic.”From Lovable’s own plan FAQ / docs framing (re-check the live selector at paste):
- Free: daily grant of 5 build credits (up to 30/month), plus small Cloud / AI feature grants described on the pricing page.
- Pro: commonly starts around $25/month for 100 credits (credit tiers scale upward; annual discounts exist — verify the dropdown live). Daily build grants continue on paid plans; credits can expire on the schedule Lovable publishes.
- Business: higher entry (commonly cited from $50/month at the 100-credit rung) for team governance — not automatically “more building power for the same credits.”
- Ownership: Lovable states you own your projects/code (subject to third-party model rights) — still put export/Git handoff in the SOW in plain English.
Credits are shared in a workspace. One chatty collaborator can empty the pool. Set per-member limits if you invite a client into the workspace for “feedback.”
v0 — UI-first React/Next path
v0.app/pricing (checked 21 Sep 2026):- Free: $0/month with $5 of included monthly credits, Deploy to Vercel, Design Mode, GitHub sync, 7 messages/day limit.
- Plus (paid entry): $30/user/month — $30 included monthly credits per user, $2 free daily credits on login, team collaboration, buy more credits as needed.
- Business: $100/user/month — training opt-out by default and team controls; included credit line matches Plus on the cards we saw — pay more for privacy defaults, not automatically for a bigger meter.
- Models are metered (v0 Mini / Pro / Max / Max Fast with published $/1M token rates on the same page).
v0 shines when the deliverable is UI + React/Next code you (or a developer) will own on GitHub and host on Vercel. It is less “chat me a whole product company” and more “generate the front-end shell fast.”
Buy/skip matrix (sheet)
| Job | Lean Bolt | Lean Lovable | Lean v0 | Skip all three |
|---|---|---|---|---|
| Agency demo / clickable concept | Strong | Strong | If React handoff matters | — |
| Founder waitlist / landing | Strong | Strong | If they want Next.js | Carrd / Webflow may be enough |
| Handoff-to-dev (GitHub) | Check export flow | Check Git sync / download rules | Natural fit | Hire the dev first |
| “Just use Webflow/Carrd” | Skip | Skip | Skip | Yes — don’t burn credits |
Export, GitHub, hosting, and who owns the code
Plain English, not legal advice: read each tool’s terms the week you sell. Confirm whether Free projects stay branded, whether paid unlocks private sharing, and whether code export/Git sync is available on the plan you quoted. Put in the SOW: who hosts the demo, who pays for Pro/credits during the build week, and what “done” means (URL + zip/repo, not “unlimited polish”).
How to scope the client SOW so credits don’t eat the fee
1. Cap revision rounds — e.g. one discovery prompt pack, two revision passes, then change-order.
2. Budget credits as a line item — “tooling up to $X included; overage billed or paused.”
3. Freeze scope — pages, auth yes/no, payments yes/no. “Also make it an EHR” is not round two.
4. Demo hosting window — 14–30 days on your account, then migrate or shut down.
5. Separate deck from app — don’t rebuild the pitch deck inside the MVP tool.
If you sell AI writing/services internationally, payment rails later matter — see Fiverr AI writing / earn in dollars. SA SME checkouts often want Instant EFT later (companion post N4) — keep this article international-first.
Practical afternoon playbook (so the demo doesn’t sprawl)
When I time-box an MVP, I run the same sequence regardless of tool:
1. One-sentence product — “Waitlist for X with email + one FAQ page.”
2. Three screens max for v1 — home, form success, admin stub or thank-you.
3. Freeze visual taste — one colour pair, one font stack, no “also try glassmorphism.”
4. Ship URL first — then one revision pass only if the timer still has minutes.
If the client wants brand guidelines mid-build, park them in a change-order. Prompt-to-app tools punish mid-flight brand kits: every restyle reburns tokens or credits. Keep a separate Canva/deck track for marketing polish if that’s what they actually wanted.
Also watch hosting adjacency. Bolt and Lovable include hosting stories on their pricing pages; v0 leans into Vercel deploy. Your SOW should say whether the client is buying a temporary demo on your account or a handoff they will bill themselves for. Mixing those two quietly is how freelancers donate unpaid DevOps hours.
This week — rebuild one past brief as a 90-minute MVP test on the free tier
Pick an old client brief (or a public sample). Time-box 90 minutes on Free: Bolt or Lovable or v0 — one tool only. Stop when the timer ends. Screenshot what shipped. Note token/credit burn. Decide which lane fits your offer before you sell it.
Soft next step
If you want the MVP builder chooser on one page (~$2–3 / R30–50), that’s the printable sheet. Everything above stays free.
Get the 1-page MVP builder chooser Optional one-page printable of this free guide. Not a guarantee of clients, savings, or income.