By the PrototypeVerdict Test Team · Pricing checked: Oct 29, 2026 · Hands-on test: not run (sign-up refused)
Not hands-on tested. We tried to create a free Claude account on Oct 29, 2026 to run our three standard test builds, and Claude’s sign-up page refused it (details below). So we haven’t built anything in Claude. Everything else on this page comes from Anthropic’s own pricing page, help center and docs, read on Oct 29, 2026, with a link next to each fact. There’s no score, and there won’t be one until our AI agents can run the same three builds every tool gets (see how we test).
Quick verdict
Our verdict
| Best for | Quick single-page tools, prototypes, mockups, landing-page designs and personal trackers made inside a chat |
| Not ideal for | A client portal or any app where outside users sign up with their own email and password, upload files or need an admin role |
| Starting price (as of Oct 29, 2026) | Free plan; Pro $20/month billed monthly or $17/month billed annually |
| Standout (documented) | Artifacts that call Claude run on Anthropic’s servers with no API key, and their AI usage counts against each user’s own plan |
| Watch out for (documented) | Saving data needs a paid plan, storage is text-only with a 20 MB cap per artifact, and people need a Claude account to open most shared artifacts |
| Our score | Not scored. We don’t score a tool until we’ve run our three test builds in it (why) |
See Claude’s plans on claude.com →
Bottom line (research only, not a test result): Claude is an AI chat assistant first. Artifacts is a feature inside that chat that turns Claude’s output into a document, design, deck, dashboard or “small interactive tool” you can edit and share. It is not a hosted app builder like Lovable, Base44 or Bolt. Anthropic’s docs describe no built-in email-and-password sign-up for your own users, no file storage for uploads, and no custom domain. On paid plans an artifact can save text data between sessions, which makes a personal invoice tracker plausible. A client intake portal, as our test prompt describes it, doesn’t fit what the docs say Artifacts can do. New to building apps with AI? Start with our beginner’s guide to vibe coding.
What is Claude Artifacts?
Anthropic describes an artifact as “anything Claude makes for you that you’d put in front of someone: a design, a deck, a document, a dashboard, or a small interactive tool.” It opens beside the conversation, and you can edit it, come back to it and share it. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them
Key points from that help article (checked Oct 29, 2026):
- Plans: artifacts are available on Free, Pro, Max, Team and Enterprise. They need the “Cloud code execution and file creation” setting (called “Create and edit files” in the newer Settings layout) turned on.
- Templates: Docs, Slides and Design templates are on every plan. Dashboards (beta) is on paid plans, and Motion (beta) is on Team and Enterprise only.
- What Claude can make: “documents, code snippets, single-page websites, images, diagrams and flowcharts, dashboards, and small interactive tools.”
- Legacy artifacts: artifacts made in a chat before September 16, 2026 keep working and can still be published, but you can’t make new ones. Several sharing rules below differ for legacy artifacts.
Source for all of the above: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them
Claude Code is a separate product. Anthropic describes it as “an agentic coding tool that reads your codebase, edits files, runs commands, and integrates with your development tools,” available in the terminal, IDEs, a desktop app and the browser. Source: https://code.claude.com/docs/en/overview. It’s included with Pro and higher, not Free. Source: https://claude.com/pricing. It works on your own code project, so it’s closer to a developer tool than to a chat-to-app builder. Claude Code can also publish what it produces in a session as an artifact: a live page at a private URL that you can share. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them. These notes cover Artifacts on claude.ai. We haven’t researched Claude Code in depth, so we make no claims here about hosting or databases in it.
Who Claude Artifacts is for
People who want a quick tool, mockup or page without leaving a chat. Claude’s Design template covers “visuals, mockups, prototypes, one-pagers, and landing pages,” and designs can be exported as a .zip, PDF, PowerPoint or standalone HTML. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them
Paid users who want a personal tracker that remembers data. On Pro, Max, Team and Enterprise, artifacts “can store data between sessions,” for things like “journals, trackers, and collaborative tools.” Source: same help article.
Teams already on Claude. On Team and Enterprise, artifacts stay inside the organization by default and can be shared with everyone there or with specific people. Source: https://support.claude.com/en/articles/9547008-share-artifacts. On Pro and higher, an artifact can also use the apps each person has connected to Claude. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them
Accountants and small firms: treat Artifacts as a place to prototype a form or a dashboard, not to run a client-facing portal. Per Anthropic’s docs, anyone opening a shared artifact needs a Claude account (except a legacy artifact published from a chat), stored data is text-only, and the person who builds an artifact decides which data is shared with everyone who opens it. Anthropic itself warns: “Before entering anything sensitive, check whether the artifact uses shared storage.” Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them. Don’t put real client or taxpayer data into a prototype.
Who should skip it: anyone who needs a real multi-user app with its own sign-up, roles, file uploads and a public URL. Look at Lovable, Base44 or Bolt, which we have tested.
Claude pricing (as of October 29, 2026)
Claude’s individual plans are Free, Pro and Max. Team and Enterprise are for organizations.
Plans
| Plan | Billed annually (per month) | Billed monthly | Usage | Relevant extras for app building |
|---|---|---|---|---|
| Free | $0 | $0 | Lower usage than Pro | Create Artifacts; Docs, Slides and Design templates. No Claude Code. Can’t store data in an artifact or share a new artifact by link |
| Pro | $17/mo ($200 billed up front) | $20/mo | 5-hour session limit plus a weekly limit | Store data in artifacts, connect apps to artifacts, share artifacts by link, Dashboards (beta), Claude Code |
| Max 5x | Not offered (monthly only) | $100/mo | 5x Pro’s per-session allowance | Everything in Pro, plus $100/month in API credits |
| Max 20x | Not offered (monthly only) | $200/mo | 20x Pro’s per-session allowance | Everything in Pro, plus $200/month in API credits |
| Team (standard seat) | $20/seat/mo | $25/seat/mo | More usage than Pro | For teams of 2 to 150; organization sharing, SSO, admin controls |
| Team (premium seat) | $100/seat/mo | $125/seat/mo | 5x a standard seat | Same as standard seat |
| Enterprise | $20/seat/mo plus usage at API rates | n/a | Usage-based | Billed annually, 20-seat minimum |
Prices in USD, as of Oct 29, 2026, read from https://claude.com/pricing (Individual and Team & Enterprise tabs). Max tier prices and “monthly only” from https://support.claude.com/en/articles/11049741-what-is-the-max-plan. Session and weekly limits from https://support.claude.com/en/articles/8325606-what-is-the-pro-plan. Artifact storage and connected apps by plan from https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them. Sharing by plan from https://support.claude.com/en/articles/9547008-share-artifacts. The pricing page says prices don’t include applicable tax.
See Claude’s plans on claude.com →
How Claude’s usage limits work
Claude doesn’t sell credits per app the way Lovable or Base44 do. Every plan has a usage limit on a rolling five-hour window, and paid plans add a weekly limit across all models. Source: https://claude.com/pricing (FAQ). Anthropic doesn’t publish a fixed message count: it depends on message length, attachments, conversation length, the model and the feature used. Source: https://support.claude.com/en/articles/8325606-what-is-the-pro-plan. Anthropic lists “artifact creation and usage” as one of the things that affect your limits. Source: https://support.claude.com/en/articles/9797557-usage-limit-best-practices. Usage on claude.ai, Claude Code and Claude Desktop all counts toward the same limit. Source: https://support.claude.com/en/articles/11647753-how-do-usage-and-length-limits-work
Paid plans can turn on usage credits to keep working past the included limit. Source: https://support.claude.com/en/articles/8325606-what-is-the-pro-plan. Anthropic’s pricing page says those credits are charged “at standard API rates” and doesn’t list a per-message price. Source: https://claude.com/pricing. The Pro plan doesn’t include API usage through the Claude Console. Source: https://support.claude.com/en/articles/8325606-what-is-the-pro-plan.
Free plan limits: Free has a session limit that resets every five hours, and the number of messages you can send “will vary based on demand.” Source: https://support.claude.com/en/articles/8114491-get-started-with-claude. The pricing page says Pro gives “at least 5x more usage per 5-hour session than Free,” but Anthropic doesn’t publish a message count for either plan. Source: https://claude.com/pricing
Is the free plan enough for our three test builds?
Not for all three, based on Anthropic’s own documentation (not a test result). Free can create artifacts, but “Store data in an artifact” is listed for Pro, Max, Team and Enterprise only. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them. Build 2 requires data that’s “still there after a page refresh,” and Builds 1 and 3 need saved submissions. A fair test would need Pro at $20/month billed monthly (as of Oct 29, 2026), which costs less than the $25/month Lovable Pro plan we tested on and the $50/month Base44 Builder plan we tested on.
Prices change. We’ll tell you when.
One email when a plan we track changes, plus a monthly roundup.
Free. Confirm by email. Unsubscribe any time. Privacy.
What happened when we tried to test Claude
On Thursday, Oct 29, 2026, at about 7:55 to 7:58 PM CT, our AI agent opened https://claude.ai/login to create a free account with our test email address. After “Continue with email,” the page showed: “Unfortunately, Claude is not available to new users right now. We’re working hard to expand our availability soon.” No account was created, so we didn’t run any builds.
We don’t know whether this applied to every new user that evening or only to our test setup. Anthropic’s help center still says anyone in a supported location can sign up for Free with an email address. Source: https://support.claude.com/en/articles/8325606-what-is-the-pro-plan. If you’re signing up yourself, you may not see this message.
How Claude Artifacts maps to our three test builds (documentation only, untested)
None of this is a test result. It’s what Anthropic’s docs say Artifacts supports, checked against what each of our standard test prompts asks for.
| What the build needs | What Anthropic’s docs say (Oct 29, 2026) | Source |
|---|---|---|
| Users sign up and log in with email and password | People using an artifact that calls Claude “sign in with their Claude account.” Anyone opening a shared artifact needs a Claude account, except a legacy artifact published from a chat. Paid plans can also invite people by email (beta), but each one still needs a Claude account with that address. The docs don’t describe a sign-up system for your own users. | Artifacts help, Share artifacts |
| Database / data saved after refresh | Paid plans only: “personal” storage (each user’s own data) or “shared” storage (everyone sees the same data). 20 MB per artifact, text only. | Artifacts help |
| File upload (PDF or image) that persists | Artifact storage “accepts text only, no images, files, or binary data.” | Artifacts help |
| Admin vs client roles | Storage is split only into personal and shared; the builder decides which is which. Sharing offers view, comment and edit access, but that controls who can change the artifact, not app roles like admin and client. The docs don’t describe user roles inside an artifact. | Artifacts help |
| Publishing a public URL | Pro and Max can share with “Anyone with the link” (Team and Enterprise too, if an owner allows it), but viewers still need a Claude account. Free can’t share a new artifact by link. Artifacts that use Claude or connected apps can’t use “Anyone with the link.” Legacy chat artifacts can be published publicly and embedded on another site. | Share artifacts |
| Code export | Design artifacts export as .zip, PDF, PowerPoint or standalone HTML. Legacy artifacts let you view, copy or download the code. The help article doesn’t say how to export code from a new chat artifact that isn’t made from a template. No GitHub sync is described. | Artifacts help |
Build 1: Client intake portal
Documented fit: poor. Clients would need Claude accounts rather than signing up in your portal, uploaded documents can’t go into artifact storage (text only), and the docs don’t describe an admin role. Shared storage would let everyone using the artifact see the same data, which is the opposite of what an intake portal needs. Test result: not run.
Build 2: Invoice tracker
Documented fit: plausible on a paid plan. A single-user tracker with personal storage matches Anthropic’s own “trackers” example, and data persists between sessions on Pro and higher. Automatic totals and an automatic Overdue status are ordinary front-end logic. On Free, storage isn’t listed, so “still there after a refresh” isn’t documented. Test result: not run.
Build 3: LedgerLight landing page
Documented fit: partial. The Design template explicitly covers landing pages and exports standalone HTML you could host elsewhere. A contact form that saves submissions would need shared storage, and anyone opening a non-legacy shared artifact needs a Claude account, so this wouldn’t work as a public marketing page as documented. Test result: not run.
What we’d test
When we can create an account, our AI agents would run the three prompts word for word on Pro (the cheapest plan with artifact storage) and log:
- Whether Claude produces a working artifact for each prompt, the time to “working,” and how many fix messages it needs (max 5 per build)
- How much of the 5-hour session limit and weekly limit the three builds use (Settings > Usage)
- Whether invoice data survives a refresh and a new session, and whether intake data lands in personal or shared storage
- What a second test account sees when it opens the shared intake portal
- What happens to a file upload in Build 1, given the text-only storage rule
- What a visitor without a Claude account sees on the landing page link
- What code we can export from a new (non-legacy) chat artifact
Our score (pending)
Not scored yet: not hands-on tested.
We haven’t given Claude a score, not even a provisional one. Every part of our score starts at 10 and loses points only for specific things written down in a test log (how we score). Without a log, most parts would be guesses.
| Part of the score (weight) | Score | What we know so far (documentation only) |
|---|---|---|
| Ease of use (25%) | Pending | Not tested. |
| Output quality (25%) | Pending | Not tested. |
| Pricing value (20%) | Pending | Pro is $20/month billed monthly. Usage limits reset every five hours and weekly rather than per app. Usage share: not tested. |
| Backend and data (15%) | Pending | Text-only storage on paid plans (20 MB per artifact); no documented file storage or custom user sign-up. Not tested. |
| Code ownership and export (15%) | Pending | HTML and .zip export documented for designs; no GitHub sync described. Not tested. |
| Overall | Pending | Added only after a hands-on test |
How Claude compares with the tools we’ve tested
| Tool | Our score | Plan we tested on | Price of that plan (monthly billing) | Built-in sign-up, database and file uploads |
|---|---|---|---|---|
| Lovable | 8.4/10 | Pro | $25/mo | Yes (tested) |
| Bolt | 8.4/10 | Free | $0 | Yes (tested) |
| Base44 | 8.1/10 | Builder | $50/mo | Yes (tested) |
| Claude (Artifacts) | Not scored | Not tested | Pro would be $20/mo | Claude accounts only; text-only storage on paid plans (documented, untested) |
Scores are from our own Oct 29, 2026 test round. Claude’s row is from Anthropic’s docs only. See also Bolt vs Lovable and Base44 vs Lovable.
Pros and cons
Pros (from documented features and pricing)
- Artifacts are on every plan, including Free
- Artifacts that call Claude run on Anthropic’s infrastructure with no API key, and each user’s AI usage counts against their own plan, not yours
- Paid plans can save data between sessions, with private per-user or shared storage
- Designs export as standalone HTML, .zip, PDF or PowerPoint
- Pro ($20/month) also includes Claude Code for people who want to work on a real codebase
Cons (from documented policies)
- Not a hosted full-stack app builder: no documented custom sign-up, user roles, file storage or custom domain
- Saving data needs a paid plan, and storage is text-only with a 20 MB cap per artifact
- Anyone opening a shared (non-legacy) artifact needs a Claude account, which rules out public-facing apps
- Unpublishing a legacy artifact permanently deletes its stored data, and it can’t be published again
- Usage limits aren’t published as fixed numbers, so cost per app is hard to predict
From our hands-on test
- Not tested. Sign-up refused on Oct 29, 2026.
Claude alternatives
- Lovable: chat-to-app builder with built-in sign-up, database, Git sync and code download, tested on Pro Oct 29, 2026.
- Base44: all-in-one builder for internal tools and portals, tested on Builder; its exported code still relies on Base44’s backend. See Base44 vs Lovable.
- Bolt.new: browser-based builder that built all three of our test apps on its free plan; see Bolt vs Lovable.
Claude FAQ
Is Claude free?
Yes. As of Oct 29, 2026, Claude has a $0 Free plan that includes creating artifacts with the Docs, Slides and Design templates. Saving data inside an artifact and Claude Code need a paid plan. Source: https://claude.com/pricing
How much does Claude Pro cost?
As of Oct 29, 2026, Pro is $20/month billed monthly, or $17/month with an annual subscription ($200 billed up front). Max is $100/month (5x) or $200/month (20x), monthly only. Sources: https://claude.com/pricing and https://support.claude.com/en/articles/11049741-what-is-the-max-plan
Can Claude Artifacts build a full app?
It can build “small interactive tools” and single-page websites, and on paid plans they can save text data. Anthropic’s docs don’t describe custom user sign-up, file storage or user roles, so it isn’t a full app builder like Lovable, Base44 or Bolt. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them
Do Claude artifacts save data?
On Pro, Max, Team and Enterprise, yes: artifacts can store data between sessions in personal or shared storage, up to 20 MB per artifact, text only. Free isn’t listed for storage. Source: https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-them
Can people without a Claude account use my artifact?
Mostly no. Anthropic says everyone needs a Claude account to open a shared artifact, even with the link. The only exception is a legacy artifact published from a chat. Source: https://support.claude.com/en/articles/9547008-share-artifacts
Have you tested Claude?
No. Claude’s sign-up page refused our new test account on Oct 29, 2026, so these research notes are based on Anthropic’s pricing page, help center and docs, checked that day. We’ll add our three test builds and a score only after a hands-on test.
Update history
- October 29, 2026: First published as research notes. Prices checked Oct 29, 2026 on claude.com/pricing and Anthropic’s help center. Not hands-on tested.