Base44 pricing in 2026: plans, credits, and what an app with Twitter (X) data really costs

The public plans on base44.com/pricing are shown in US dollars per month, with a Yearly / Monthly toggle. Yearly is the default view and is labeled "SAVE 20%"; the FAQ explains that the discount applies to the monthly price times twelve, charged once a year.

What each tier adds. Free gives every core feature (AI building, backend and database, auth, analytics) for up to 5 apps. Starter lifts the app limit to unlimited and adds in-app code edits and a custom domain. Builder, marked "Popular", adds AI model selection and GitHub integration. Pro adds early access to beta features. Elite adds premium support and the largest credit pools. Enterprise adds custom credits, a dedicated architect, SSO, advanced security controls and an account team.

Yearly-only extras. On the yearly toggle, paid plans list a free domain for one year, and Builder and up list 25 credits to share with a friend. Those two lines disappear when you switch the toggle to Monthly, so they are part of the yearly deal rather than the plan itself.

Payment and changes. Base44 takes Visa, Mastercard, American Express and Discover; Enterprise can pay by invoice or ACH. You can upgrade, downgrade or cancel at any time, changes apply at the next billing cycle, and the FAQ says apps stay live. Education and nonprofit pricing is handled case by case, and founders can apply to Base44 for Startups for a free annual license and credits.

Base44's credits documentation (docs.base44.com/Account-and-billing/Credits) splits usage into two pools, and the split is the most important thing to understand before choosing a plan.

Message credits are spent when you prompt the AI to plan, change or fix the app. There is no fixed price per message; the docs give rough ranges and say the real figure is calculated after the message runs. Manual visual edits in the editor (dragging elements, changing layouts, editing text) use no credits.

Integration credits are spent by the running app when it uses Base44's built-in services. The docs list SendEmail at about 1 credit per email (about 2 from a custom domain), GenerateImage at about 1 per image, GenerateVideo at about 5 per second of video, invokeLLM on the Automatic model at about 1 per call, in-app agent messages from about 3 credits each on Automatic and several times more on larger models, workflows at about 1 credit per 10 steps, and backend functions at about 1 credit per 100 invocations. The docs also state that integration credits are not used when you call an external service with your own API key.

Credits per dollar. On monthly billing every plan from Starter to Pro buys 5 message credits per dollar ($0.20 each); yearly billing makes that 6.25 per dollar ($0.16 each). Elite is the only tier with a better ratio: 6 per dollar monthly and 7.5 yearly (about $0.17 and $0.13 each). Integration credits scale faster than price: 100 per dollar on Starter, 200 on Builder and Pro, 250 on Elite (monthly billing).

At the docs' own estimates, Starter's 100 message credits are roughly 100 small feature changes or about 25 to 33 app-wide rewrites a month, and Builder's 250 are two and a half times that. That is the real sizing question: how many prompts you send, not how many apps you own.

Reset date. Paid message and integration credits reset once a month on the day you subscribed (subscribe on the 5th, reset on the 5th). The docs say credits you have not used expire at the reset. On Free, your personal workspace gets up to 5 message credits a day and 25 a month; the daily allowance resets at midnight in the owner's time zone, and other workspaces on Free get no free message credits.

Running out. When you hit the limit, actions that use credits pause until the next reset. Your app stays up, but AI building (message credits) or built-in services (integration credits) stop until then, unless you upgrade from the billing dashboard.

Upgrading mid-cycle. An upgrade gives the new plan's extra credits right away. Moving from a monthly plan to a higher monthly plan charges the full price difference, and your unused credits carry over for one more cycle (up to one month of credits) before expiring. Any upgrade that involves a yearly plan is prorated, and unused credits expire at the next reset as usual.

No top-up store on the pricing page. The pricing page lists no credit packs; the documented paths to more credits are a higher plan, Enterprise custom credits, or a coupon code redeemed on the usage page. Base44 gift cards are no longer sold, according to the same docs page.

Workspace-wide pool. Credits are shared across every app in a workspace, and owners on plans with member limits can cap how many credits each member uses. The usage page breaks spend down by app, member, source (workflows, backend functions, agents) and branch.

Every AI app builder meters something different, so a straight price comparison only goes so far. These are the entry paid plans as each vendor's own pricing page or docs showed them when we checked this year.

Base44's distinctive line item is the second pool. Lovable, Replit and Bolt price the building work; Base44 also prices what the live app does with its built-in email, image, LLM and agent services. If your app sends a lot of email or runs in-app agents, compare integration credits, not just the headline price.

Rollover is where Base44 is strict. Bolt says unused tokens roll over to the next month and Lovable's monthly credits stay valid for two months; Base44's docs say unused credits expire at each reset. A public Base44 feature request to roll unused credits over exists, but it is a request, not a policy.

For the Lovable side in full (every Pro and Business credit tier, top-ups, Cloud costs), see our Lovable pricing breakdown linked below. Replit Pro was listed at $100 a month ($90 yearly) with $100 of credits, and Bolt Teams at $30 per member a month.

We pulled 883 recent X posts matching Base44 plus a money word (credits, pricing, plan, subscription, cost, pay and similar), kept the 755 posted since the @Base44 account was created in September 2024 that actually contain a pricing term, and tagged them with transparent keyword rules. They come from 520 distinct authors; 97.9% are in English, 26.2% are replies, and 33.6% tag @Base44 directly.

Money talk about Base44 is mostly about the company, not the bill. The largest single theme is Wix: 22.9% of the posts mention Wix, split between $WIX stock commentary (12.3%) and retellings of the solo-founder-to-acquisition story (11.7%). For a buyer reading X to decide on a plan, roughly a fifth of the search results are noise.

Complaints are real but small. 13.8% of posts carry a cost complaint of some kind (the word expensive, burned credits, fix loops, refunds), against 2.8% that call Base44 worth it or good value, about 5.0 to 1. Posts saying credits ran out are only 1.6% (12 posts), and refund or cancel talk 0.9%. Read the expensive row with care: the keyword also fires on builders pitching Base44 as a way to replace expensive tools.

Free credits and the second pool get attention. Free and daily credits are 10.5% of posts, and backend or integration credits 10.7%; the integration-credit posts have the highest median views of the product themes (995), a sign that the two-pool model is what people need explained.

Volume is rising. 78.4% of the on-topic posts are from 2026, and the most recent partial quarter already has 134 posts, against 69 in the same quarter of 2025.

26.2% of the posts name at least one other builder or coding tool. Lovable is the default comparison by a wide margin.

Lovable first, then the coding agents. Lovable appears in 16.2% of posts, Claude (mostly Claude Code) in 10.5%, Replit in 9.8% and Cursor in 8.1%. Bolt (6.5%) and v0 (3.7%) trail. The pattern matches the buying question: people choose between Base44 and Lovable as no-code builders, and between a builder and a coding agent once the app gets serious.

If you are weighing that second step, our Claude Code pricing page and the Cursor vs Claude Code comparison cover the agent side of the bill.

The official account (@Base44, created in 2024) had 39,571 followers when we collected. Its 180 most recent posts span 194 days, about 0.93 posts a day.

Pricing posts outperform the average. About one post in five (18.3%) touches credits, plans or free offers, and those posts get a median of 54 likes against 43 for the account overall. The official feed almost never mentions Wix (1 post), while Wix is the biggest money topic in the public conversation.

If you want to track a vendor's pricing announcements like these yourself, the same /twitter/user/last_tweets call used for this study works on any public account.

Base44 apps can call outside APIs from backend functions: Deno serverless functions that run server-side, read secrets at runtime and are invoked from the frontend with base44.functions.invoke() (docs.base44.com/developers/backend/resources/backend-functions/overview). That is the right place for an X data call, because the API key never reaches the browser.

Step 1: store the key. Create a key in the TwitterAPI.io dashboard and set it as a Base44 secret with the CLI: base44 secrets set TWITTERAPI_KEY=your_key. Base44's docs note that a function must call secrets.get() inside the handler; a read at module load returns undefined.

Step 2: the function. The function below checks the signed-in user, calls /twitter/tweet/advanced_search on https://api.twitterapi.io with the X-API-Key header, follows has_next_page and next_cursor for a few pages, and returns slim tweet objects (text, url, createdAt, likeCount, viewCount, author.userName). Search results come back in a top-level tweets array.

Step 3: call it from the app. In the frontend, base44.functions.invoke("xSearch", { query }) returns the JSON. You can also prompt Base44's AI to build the UI around that function; the prompt costs message credits, but the X call itself does not.

What it costs in Base44 credits. Because the function calls an external service with your own key, the docs say no integration credits are used for the call itself; the function invocation costs about 1 integration credit per 100 runs. A dashboard that refreshes every 5 minutes runs 8,640 times a month, about 86 integration credits, well inside Starter's 2,000.

Other endpoints. Swap the path for /twitter/user/last_tweets (an account's timeline, under data.tweets), /twitter/user/mentions (posts mentioning a handle) or /twitter/user/info (a profile). Reads need no X account. If you build with a coding agent instead of Base44, our hosted MCP server exposes the same data as tools.

The X data arrives as a separate line item from the X data provider. TwitterAPI.io bills 15 credits per tweet returned, which is $0.15 per 1,000 tweets, with a minimum of 15 credits per request; $1 buys 100,000 credits. A full 20-tweet page costs $0.003, and a search that finds nothing costs the $0.00015 minimum.

The official X API pay-per-usage rate card at docs.x.com/x-api/getting-started/pricing lists $0.005 per post read and $0.010 per user read, capped at 3 million post reads a month before Enterprise. Per 1,000 posts that is $5 against $0.15, about 33 times more.

Read the refresh rows as the worst case for the floor: 8,640 checks a month that each found nothing add up to $1.30 at the $0.00015 minimum; checks that return tweets are billed per tweet instead. A dashboard that shows about 3,000 new tweets a month therefore lands under $2 of X data, next to a Base44 Starter plan at $20 a month ($16 yearly).

For scale, the whole X study on this page (883 search results, 180 official-account posts and one profile lookup, 56 requests) cost 15,963 credits, about $0.16.

Paid accounts get 20 requests per second (0.2 before the first payment). New accounts that pass the automatic bot check get up to $1 of free credits (at most 100,000 credits, no card, one per person or network, expiring after 31 days); purchased credits never expire. Monthly subscriptions start at Starter $29 a month; see the pricing page for details.

Question. When people on X talk about Base44 pricing and credits, what do they say, which tools do they compare it with, and what does the official account post about plans?

Sources and query. /twitter/tweet/advanced_search with Latest ordering on (base44 OR @base44) plus money words (credits, credit, pricing, price, plan, subscription, expensive, cost, paid, pay, a dollar sign) excluding retweets, 45 pages; /twitter/user/last_tweets for @Base44, 10 pages; /twitter/user/info for @Base44, 1 request. 56 requests in total, collected in 2026.

Sample. 883 search posts returned; 6 posted before September 2024 were dropped as unrelated uses of the string; 755 contained a pricing term and form the sample, from 520 distinct authors. The official sample is 180 posts.

Method. Each post is tagged by keyword rules published in analyze.py and results.json in our study folder; a post can carry several themes. Shares are of the on-topic sample. Likes and views are medians.

What the sample cannot show. It is the most recent matching posts on X, not a random sample of Base44 users. Keyword rules miss sarcasm and catch unrelated senses of words like expensive. Posts in other languages that never use the English money words are mostly missed. Nothing here measures how many credits real apps spend.

Privacy. Aggregates only. Raw search pages were kept out of the repository because their authors are private individuals; the de-identified rows hold no IDs, URLs, text or handles. Only the official @Base44 organization account is named.