Is Claude Free? What the Free Plan Includes (2026)

Is Claude free? Yes. Anthropic's Claude has a Free plan that costs $0, and it covers chat on the web, desktop and mobile with limited usage every 5 hours and every week. Claude Code, the Opus models and Research are not part of Free. You need Pro ($20 a month) or higher for those.

By Gary, Greenlight My App · Updated Oct 31, 2026 · How we checked this: On Oct 31, 2026 at 6:53 AM ET we captured Anthropic's pricing page and read its plan cards, feature table and FAQ. We also read Anthropic's privacy articles, the Claude Code setup docs and OpenAI's ChatGPT pricing page. We didn't create a Free account or run Claude Code for this page. The key example below is a local text search on two demo files with a fake key, not a Greenlight My App scan. Drafted with AI tools and reviewed by a person before publishing.

People ask "is Claude free" for two different reasons. Some want a free AI chat for writing, study or research. Others are building an app with AI help and want to know what they can do before paying. This page answers both, and then covers the part most guides skip: what not to paste into a free chat.

Is Claude free? The short answer

Is Claude free to use? Yes, with a Free plan. On Anthropic's pricing page (external), checked Oct 31, 2026, the Free card reads "$0" and "Free for everyone," with the tagline "Try Claude." The plan table lists the credit card payment option as "n/a" for Free, since there is nothing to pay.

The Free plan (what many people call the Claude free tier) is a real plan, not a short demo that expires. But it is limited in two ways:

1. Usage limits. The Free card says "Limited usage every 5 hours and every week." When you hit the cap, you wait for it to reset or upgrade.
2. Feature limits. Some tools only exist on paid plans. The biggest ones for builders are Claude Code and the Opus models.

If you only want to know about the coding tool, we cover that separately in Is Claude Code free?. Short version: Claude Code isn't included in Free.

What does the Claude Free plan include?

The Free card on claude.com/pricing lists these features, word for word, as of Oct 31, 2026:

• "Limited usage every 5 hours and every week"
• "Chat on web, desktop, and mobile"
• "Search the web, create files, and run code"
• "Memory across conversations"
• "Connect your apps and tools"
• "Create Artifacts"
• "Claude Design, Slides, Docs"

The full feature table on the same page adds more detail. Here is what it shows for Free compared with Pro:

Prices as of Oct 31, 2026. Check the official page before you rely on them.

What can you actually do with that? Quite a lot for free:

• Ask questions, draft emails and summarize documents.
• Search the web from inside the chat.
• Have Claude write and run small bits of code to analyze a file or make a chart.
• Keep up to 5 Projects for separate topics.
• Build a simple Artifact, like a calculator or a one-page mockup.

Worked example. Say you're planning a small landing page. On Free, you could ask Claude to draft the copy, build a rough HTML Artifact to preview it, and search the web for examples. What you can't do on Free is hand the work to Claude Code to edit the files in your project folder. That needs Pro or higher.

What is the free Claude limit?

There is no published message number. Anthropic's pricing FAQ says "there's no fixed message count," because the limit depends on what you do. The footnote under the plan cards says how fast you reach a limit depends on the model you use, how long your chats are and similar factors.

What Anthropic does publish:

• Free has "Limited" usage in both the "Limit every 5 hours" and "Limit every week" rows of the plan table.
• Pro is the 1x baseline in that table. Max is 5x and Ultra is 10x Pro weekly usage.
• The pricing FAQ says: "Pro gives you at least 5x more usage per 5-hour session than Free."
• The FAQ also says "You can see where you stand anytime in Settings > Usage."

How do you make the free limit last longer? These tips follow from how Anthropic says limits are counted (model, chat length and tools):

1. Start a new chat for a new topic. Long chats carry more history, which counts against your usage.
2. Pick the lighter model for easy tasks. Free includes Sonnet and Haiku. Use the lighter one for quick rewrites.
3. Paste only what's needed. Share the relevant part of a document, not the whole thing.
4. Check Settings > Usage before you start a long piece of work.

We didn't run a Free account to measure where the cap kicks in, and we don't quote message counts. Anyone who gives you an exact number for Free is guessing, because Anthropic says there isn't one.

What's not free: Claude Code, Opus, Research and more

These items are marked as not included on Free in the Oct 31, 2026 plan table:

Prices as of Oct 31, 2026. Check the official page before you rely on them.

Claude Code. The Claude Code docs say it plainly: "Claude Code requires a Pro, Max, Ultra, Team, Enterprise, or Console account. The free claude.ai plan does not include Claude Code access." A Console account means paying for API usage. For the full breakdown, see Is Claude Code free? and Claude Code pricing.

The Claude API. The developer API is separate from the chat plans. It's billed per token through the Claude Console. Our Claude API pricing guide covers it. A Free chat plan doesn't give you API access or API credits.

Claude Free vs Pro vs Max: what changes?

Here are the individual plans side by side, from claude.com/pricing on Oct 31, 2026:

Prices as of Oct 31, 2026. Check the official page before you rely on them.

Note the Ultra tier. Older guides, including some of ours from earlier in October, talk about Free, Pro and Max only. The Oct 31, 2026 pricing page also lists Ultra at $200 a month. Check the live page before you decide.

Worked example: should you upgrade? Say you use Claude 30 minutes a day for writing and you rarely see a limit notice. Free is probably enough. Now say you want Claude to edit code in your project folder every evening. Free can't do that at all, because Claude Code needs Pro. At $20 a month, that's $240 a year, or $200 if you pay yearly up front. Whether that's worth it depends on how much time it saves you.

Is Claude AI free to use for work or a startup?

You can use the Free plan for work tasks like drafting, research and planning. But a few things matter more at work:

• Training is opt-out on Free. The plan table shows "Model training: Opt-out" on every individual plan, including Free. That means it's on unless you turn it off. More on that below.
• No admin controls. The individual plans have no single sign-on, central billing or audit logs in the plan table. Anthropic lists Team and Enterprise plans separately for businesses.
• Limits can interrupt you. If you hit the 5-hour cap in the middle of a deadline, you wait.

For a solo founder or a side project, Free is a fine place to start. For a team that handles customer data, read Anthropic's commercial terms and the Team or Enterprise plan details before anyone pastes real data in.

Is it safe to paste code or keys into free Claude?

Pasting code is common and usually fine. Pasting secrets is not. Here's what Anthropic's own pages say.

Training. Anthropic's privacy article on model training covers "Claude Free, Pro, Max and when accounts from those plans use Claude Code." It says chats are used to improve models if "You choose to allow us to use your chats and coding sessions to improve Claude," if they're flagged for safety review, or if you opt in another way. It also says: "Your Incognito chats are not used to improve Claude, even if you have enabled Model Improvement in your Privacy Settings."

How to turn training off. Anthropic's privacy settings article (external) says to open Settings, then Privacy, and toggle the setting off. It adds: "If you decide to turn off the model training setting, we will not use any new chats and coding sessions you have with Claude for future model training." Data already used in training stays there.

Secrets. Anthropic's help article on sensitive data recommends "being thoughtful about sharing highly sensitive personal details such as" financial information, health records and "Passwords or private login credentials." An API key is a login credential for your account. Treat it like a password.

A practical rule set:

1. Paste code, not secrets. Replace any key with a placeholder like sk-ant-XXXX before you paste.
2. Don't paste .env files. Paste the variable names only.
3. Turn off training in Settings > Privacy if you share work code.
4. Use Incognito chats for anything sensitive. Anthropic says those aren't used for training.
5. If you did paste a real key, rotate it in the provider's console. Deleting the chat doesn't take the key back.

What if a free AI chat put my API key in the frontend?

This is the bigger risk for people building apps with free AI tools. A chat assistant may hand you code that calls the Anthropic API straight from the browser, with the key in your JavaScript. Anyone who opens your site can read it.

Here is the risky pattern, from our demo file (the key is a fake placeholder):

<script>
  // risky: this key ships to every visitor's browser
  const ANTHROPIC_API_KEY = "sk-ant-XXXX";
  fetch("https://api.anthropic.com/v1/messages", {
    method: "POST",
    headers: { "x-api-key": ANTHROPIC_API_KEY, "anthropic-version": "2023-06-01" },
  });
</script>

And the safer pattern: the browser calls your own server route, and the server reads the key from an environment variable.

// browser: no key here
fetch("/api/chat", {
  method: "POST",
  headers: { "content-type": "application/json" },
  body: JSON.stringify({ message: "hi" }),
});

// app/api/chat/route.ts (server only)
export async function POST(req: Request) {
  const { message } = await req.json();
  const r = await fetch("https://api.anthropic.com/v1/messages", {
    method: "POST",
    headers: {
      "x-api-key": process.env.ANTHROPIC_API_KEY ?? "",
      "anthropic-version": "2023-06-01",
      "content-type": "application/json",
    },
    body: JSON.stringify({
      model: "YOUR_MODEL_ID", // pick a current model from Anthropic's docs
      max_tokens: 256,
      messages: [{ role: "user", content: message }],
    }),
  });
  return new Response(await r.text(), { status: r.ok ? 200 : 502 });
}

How we checked the demo. We ran a simple text search for the sk-ant- prefix on both demo files on Oct 31, 2026. It flagged the risky file and found nothing in the fixed one. This was a local search on files, not a scan of a live site.

You can run the same kind of search for Anthropic and OpenAI key prefixes on your own build output before you deploy:

grep -rnE "sk-ant-|sk-proj-" dist/ .next/static/ 2>/dev/null

Once your app is live, run a free scan with Greenlight My App. It checks your site's public JavaScript for full-length secret keys, including Anthropic sk-ant- keys, plus security headers, exposed .env and .git files and public source maps. It looks for real key formats, so a short placeholder like sk-ant-XXXX won't trigger it. For more on handling Claude keys, read Claude API key security. To check your headers on their own, try the security headers checker.

Claude vs ChatGPT free plans

Compared on Oct 31, 2026; based on Anthropic's and OpenAI's public sites (claude.com/pricing and chatgpt.com/pricing). Each vendor describes its plan differently, so this table uses their own words.

Prices as of Oct 31, 2026. Check the official page before you rely on them.

What's the difference between Claude and ChatGPT free plans? On paper, ChatGPT Free promises unlimited text chats and limited access to many tools, including Codex. Claude Free has capped usage but a much larger listed context window, plus memory, connectors and up to 5 Projects. For a longer comparison of the two assistants, see Claude vs ChatGPT.

Why do people switch from ChatGPT to Claude? Common reasons are writing style, long documents and coding help. Those are opinions, not facts we tested. The fair test is to try both free plans on the same three tasks you actually do and compare.

What if Claude isn't working?

Sometimes a "limit" is really an outage. If Claude stops replying, check Anthropic's status page first. Our Is Claude down? guide explains how to tell an outage from a usage cap or a local problem.

FAQ

Is Claude free? Yes. Claude has a $0 Free plan with chat on web, desktop and mobile, web search, files and code runs. Usage is limited every 5 hours and every week. Checked Oct 31, 2026.

Can I use Claude for free? Yes. Create an account at claude.ai and you start on the Free plan, which is $0. Claude Code, Opus and Research need a paid plan.

What is the free Claude limit? Anthropic says there's no fixed message count. Free has limited usage every 5 hours and every week, depending on model and chat length. Check Settings > Usage.

Does Claude Free include Claude Code? No. Claude Code's docs say it needs a Pro, Max, Ultra, Team, Enterprise or Console account, and the free plan doesn't include it. Pro is $20 a month.

What's the difference between Claude and ChatGPT? Both are AI assistants from different companies. On free plans, ChatGPT lists unlimited text chats; Claude lists capped usage with a bigger context window. Compared Oct 31, 2026.

Why switch from ChatGPT to Claude? People often cite writing style, long documents or coding help. Try both free plans on the same tasks before you decide, and compare what you actually get.

Does Anthropic train on Claude Free chats? Training is opt-out on Free and other individual plans. Turn it off in Settings > Privacy. Anthropic says Incognito chats aren't used for training.

Is Claude Pro worth it? If you need Claude Code, Opus, Research or keep hitting limits, maybe. Pro is $20 a month or $17 a month billed yearly as of Oct 31, 2026.

Free AI tools wrote your app? Check it free: run a free scan with Greenlight My App to check your live site's public code for exposed API keys and missing security headers.