Claude AI Pricing 2026: All Plans & Costs in USD + AUD (Free, Pro, Max, Team, Enterprise)

Updated June 2026 — every figure below is checked against Anthropic's current published rates. The big changes since our last review: Claude's API got dramatically cheaper at the top end (Opus 4.7 now runs USD $5/$25 per million tokens, down from $15/$75 on the old Opus 4), the consumer line-up is now five tiers (Free, Pro, Max, Team, Enterprise), and Pro bundles Claude Code and Claude Cowork. Australian buyers should note Anthropic adds 10% GST at checkout unless you register an ABN (covered below).

Quick answer — how much does Claude cost? Claude has a free tier, then Pro at USD $20/month ($17 on annual billing), Max from USD $100/month (up to $200 for 20× usage), and Team at USD $25/seat/month ($20 annual, 5-seat minimum). Enterprise is custom-quoted. API access is billed per token, from USD $1 per million input tokens (Haiku 4.5) to $5 (Opus 4.7). In Australian dollars that lands at roughly A$30/month for Pro, ~A$150–$300 for Max, and ~A$38/seat for Team, plus 10% GST unless you register an ABN. Full breakdown of every plan — and how to pick the right one — below.

Before you commit to an AI subscription, you need to know exactly what you are paying for. Claude AI pricing has evolved significantly since Anthropic first launched the platform, and in 2026 the options range from a generous free tier to enterprise contracts that scale across entire organisations.

We have helped dozens of businesses integrate Claude into their workflows. The most common question we hear is not "is Claude good?" — it is "which plan should I actually be on, and am I overpaying?"

This guide breaks down every Claude AI pricing tier, covers API costs for developers, and gives you a practical framework for choosing the right plan.

How Much Does Claude Pro Cost in Australia (2026)?

Claude Pro costs around AUD $30 per user per month (USD $20 monthly, or USD $17 on annual billing — roughly A$26 — both billed in USD). Claude Max runs from ~A$150/month (USD $100, 5× Pro usage) to ~A$300/month (USD $200, 20×). Claude Team is ~A$38 per seat per month (USD $25 monthly, or USD $20 on annual billing) with a 5-seat minimum. Claude Enterprise is custom-quoted — the published structure is roughly USD $20 per seat plus usage billed at API rates, with volume discounts. Anthropic bills in USD and adds 10% Australian GST by default; if you register your ABN under Settings → Billing in Claude, GST is removed from future invoices and you account for it via the reverse charge in your BAS (usually nil net effect when the spend is fully creditable).

The full breakdown by plan, including hidden costs and the right tier by team size, is below.

Claude AI Pricing Overview (2026)

Claude offers five plans: Free, Pro, Max, Team, and Enterprise. API access is priced separately on a per-token basis. The right choice depends on your team size, usage volume, and whether you need API access or just the chat interface.

Here is a quick summary before we get into the details.

Free Tier: What You Get for $0

The Claude AI free tier gives you access to the chat interface at claude.ai with a daily message cap. It is genuinely useful for individual exploration and light tasks.

What is included:

• Access to Claude Sonnet and Haiku models
• Limited access to Claude Opus (the most capable model)
• File uploads and document analysis
• 200K token context window
• Basic conversation history

What to watch for:

• The daily message limit resets every 24 hours, but during peak times you may hit the cap quickly
• You will get bumped to slower queues when demand is high
• No team features — it is a single-user experience
• Conversations may be used for model improvement unless you opt out

The free tier is fine for testing Claude, occasional personal use, or evaluating whether it fits your workflow. It is not sufficient for any kind of regular business use.

Claude AI Pro Pricing: $20/Month

The Pro plan is where most individual professionals start. At USD $20 per user per month (or $17 on annual billing), it unlocks significantly higher usage limits and full access to every Claude model.

What you get over Free:

• Roughly 5x the message allowance of the free tier
• Full, unrestricted access to Claude Opus for complex reasoning tasks
• Claude Code and Claude Cowork — the agentic coding and work tools, now bundled into Pro
• Priority access during high-traffic periods
• Early access to new features and models
• Unlimited projects and extended conversation memory

Is Claude AI Pro worth $20/month?

For anyone using Claude more than a few times per week for real work — writing, analysis, coding, research — the Pro plan pays for itself almost immediately. A single hour saved per week on document analysis or drafting easily justifies the cost.

Where the Pro plan falls short is for teams. There are no shared workspaces, no admin controls, and no centralised billing. If you have more than a few people using Claude, look at the Team plan below — it adds exactly those things from 5 seats up. If you are hitting Pro's usage ceiling as an individual (heavy Claude Code use, long research sessions), Max is the upgrade, not Enterprise.

Claude Max Pricing: $100–$200/Month

Max is for individuals who hit Pro's limits — heavy Claude Code users, people running long research sessions, or anyone who lives in Claude all day. It is the same single-user experience as Pro with a much higher ceiling.

• Max 5× — USD $100/month, roughly 5× Pro's usage (~A$150/month).
• Max 20× — USD $200/month, roughly 20× Pro's usage, with the highest output limits and priority access during peak traffic (~A$300/month).

There are no team or admin features in Max — it is a personal plan. If you need more than one seat, Team is the right move even if each person is a heavy user.

Deciding between the tiers, or whether Max is justified at all? We ran the Australian break-even maths in Is Claude Max worth it? — the short version is that Max 5× has to recover about 75 minutes a month to pay for itself.

Claude Team Pricing: $25/Seat/Month

Team is the plan built for businesses, and the one most Australian SMBs should be looking at first. It starts at 5 seats and scales to 150.

• Standard seat — USD $25/seat/month, or USD $20/seat/month on annual billing (~A$30–$38/seat).
• Premium seat — USD $125/seat/month ($100 annual) for members who need much higher usage — effectively Max-level limits inside the team.
• 5-seat minimum, up to 150 people.

What Team adds over Pro:

• Shared projects and a central workspace
• Admin console with user management and centralised billing
• SSO / SAML
• Enterprise search across your team's connected tools
• Claude Code and Cowork for every seat
• No training on your data by default

For a 5–150 person business that wants everyone on Claude with a single invoice and basic controls, Team is the sweet spot. You only need Enterprise when you require audit logs, SCIM provisioning, compliance tooling, or a contractual SLA. For the full business-plan decision — including whether Claude trains on your data and where it is stored for Australian teams — see our Claude Team vs Enterprise comparison.

Enterprise Pricing: Custom Plans for Organisations

Claude Enterprise is designed for companies that need Claude embedded across teams with governance, security, and scale.

Pricing is custom. The published starting structure is around USD $20 per seat plus usage billed at API rates, with volume discounts — but real Enterprise deals are quoted by Anthropic sales against your seat count, usage, and compliance requirements. There is no flat per-seat list price at this tier.

What Enterprise includes:

• Everything in Team
• Usage that scales at API rates rather than a fixed seat cap
• Role-based access control and SCIM user provisioning
• Audit logs and the Compliance API
• Custom data retention and zero-data-retention options (your data is never used for training)
• IP allowlisting and a HIPAA-ready offering
• Dedicated support and custom SLAs

Who should consider Enterprise:

• Companies with 10+ Claude users
• Regulated industries (finance, healthcare, legal) that need data controls
• Teams that need audit trails and compliance reporting
• Organisations that require guaranteed uptime SLAs

To get Enterprise pricing, you need to contact Anthropic's sales team directly. We can also help you navigate this process — more on that below.

How Much Does Claude Code Cost? (2026)

Claude Code has no separate price — it is bundled into every paid Claude subscription. If you are on Pro (USD $20/month, ~A$30), Max, Team, or Enterprise, both Claude Code and Claude Cowork are included at no extra cost. So for most people, the price of Claude Code is the price of a Pro seat.

What actually varies is how much Claude Code you can run before you hit your plan's usage limit:

Prefer to pay per use? You can also run Claude Code against the API with your own key, billed at standard per-token rates (see the table below) instead of a flat subscription. That suits occasional or programmatic use where a monthly seat would sit idle — but most individuals are better off on Pro or Max once they use Claude Code daily.

In short: there is no standalone Claude Code price — it rides on your Pro, Max, or Team subscription, or on API usage if you bring your own key.

Claude API Pricing: Per-Token Costs

If your team is building applications with Claude — not just chatting with it — you need the API. API pricing is completely separate from the subscription plans and is based on token usage.

Here are the current Claude API rates as of June 2026 (USD, per million tokens):

Note the big shift from a year ago: Opus dropped from $15/$75 to $5/$25, so the most capable model is now roughly a third of its old cost. Sonnet held at $3/$15; Haiku moved to $1/$5 with a much stronger model.

What does this mean in practical terms?

• A typical business email (~500 words) costs roughly $0.001–$0.002 to process with Sonnet
• Analysing a 50-page contract with Opus 4.7 costs approximately $0.10–$0.25
• Running a customer support bot on Haiku 4.5 that handles 1,000 conversations per day costs roughly $4–$8/day

API pricing scales linearly. There are no minimum commitments, and you only pay for what you use. The Batch API halves both input and output costs (50% off) for non-time-sensitive work, and Anthropic offers further volume discounts through committed-use agreements at scale.

Prompt caching can significantly reduce costs for repetitive workloads. If your application sends the same system prompt or reference documents with every request, cache reads are billed at a 90% discount (0.1× the input rate). This is a major cost saver for production applications.

Need help figuring out the most cost-effective way to use Claude for your business? We help companies optimise their AI spend.

Which Claude AI Plan Is Right for You?

Here is a simple decision framework:

Choose Free if:

• You are evaluating Claude for the first time
• You use AI tools occasionally, not daily
• You are an individual with no team requirements

Choose Pro if:

• You use Claude multiple times per day for real work
• You need reliable access to Claude Opus
• You are an individual professional or part of a small team (under 5 people)
• You want priority access and do not want to hit usage caps

Choose Enterprise if:

• You have 10+ people who need Claude access
• You operate in a regulated industry
• You need SSO, admin controls, or data retention policies
• You require an SLA for uptime and support
• You want zero data retention guarantees

Choose API if:

• You are building Claude into a product or internal tool
• You need programmatic access for automation
• You want to control costs at the per-request level
• You are running high-volume workloads (customer support, document processing)

Many businesses end up using a combination. Pro subscriptions for knowledge workers who interact with Claude directly, plus API access for developers building internal tools and automations.

Hidden Costs and What to Watch For

Claude AI pricing is straightforward compared to many SaaS tools, but there are a few things that catch businesses off guard.

1. API costs can spike unexpectedly. If you are building with the API, a poorly optimised prompt or a sudden traffic surge can run up your bill quickly. Always set spending limits in the Anthropic console and monitor usage weekly.

2. Opus costs more than you need for routine tasks. The 4.7 price cut narrowed the gap, but Opus still costs about 1.7x Sonnet per token. Many tasks that seem like they need Opus run just fine on Sonnet. Default to Sonnet and only escalate to Opus when quality demands it.

3. Long context windows cost more. Sending 200K tokens of context with every request is technically possible, but it is also expensive. Be selective about what context you include.

4. Per-user pricing adds up for large teams. At $20/user/month for Pro, a 50-person team is spending $12,000/year. At that scale, Enterprise pricing with volume discounts almost always makes more sense.

5. No rollover on unused messages. Pro plan message allowances reset monthly. You cannot bank unused capacity for a busy month.

6. The GST you cannot claim back. Anthropic adds 10% Australian GST by default and, because offshore suppliers on simplified GST registration hold an ARN rather than an ABN, that GST generally does not support an input tax credit. Registering your ABN in billing settings removes the charge entirely — but only going forward. See Does Anthropic charge GST in Australia? for the full treatment.

7. Usage limits are not message counts. Anthropic meters on tokens, not messages, across a rolling five-hour session window plus a weekly cap — which is why two people on identical plans can have a tenfold difference in what they get. Claude usage limits explained covers what actually burns your allowance and how to stop hitting limits without upgrading.

Cost Optimisation Tips

Based on our experience building Claude-powered solutions for clients, here are the most effective ways to reduce your Claude AI costs:

• Match the model to the task. Use Haiku for classification and simple extraction. Use Sonnet for most business tasks. Reserve Opus for complex reasoning, multi-step analysis, and high-stakes outputs.
• Implement prompt caching. For API users, cached prompts can cut costs by up to 90% on the system prompt portion of each request.
• Trim your context. Do not send entire documents when you only need specific sections. Pre-process inputs to extract relevant content before sending it to Claude.
• Batch similar requests. If you are processing many similar documents, batch them where possible to reduce overhead.
• Set hard spending limits. Anthropic's console lets you set monthly spend caps on your API key. Use them. A runaway script on a Friday night can be expensive.
• Monitor and review weekly. Track which models, features, and team members are driving the most usage. Optimise based on data, not guesses.

Frequently Asked Questions

How much does Claude AI cost?

Claude offers five plans: a free tier, Pro at USD $20/user/month, Max from USD $100/month, Team from USD $25/seat/month (5-seat minimum), and custom Enterprise pricing. API access is separate, starting at $1 per million input tokens for Haiku 4.5 and $5 per million for Opus 4.7.

How much does Claude Code cost?

Claude Code has no separate price — it is bundled into every paid Claude subscription. If you have Pro (USD $20/month, about A$30), Max, Team, or Enterprise, Claude Code and Claude Cowork are included at no extra cost. Heavy, all-day Claude Code users usually move from Pro to Max (USD $100–$200/month) for higher usage limits. You can alternatively run Claude Code against the API with your own key, billed per token at standard API rates.

How much does Claude Team cost?

Claude Team costs USD $25 per seat per month (about A$38), or USD $20 per seat on annual billing, with a 5-seat minimum and up to 150 seats. It adds shared projects, an admin console, centralised billing, SSO, and Claude Code and Cowork for every seat. A premium seat at USD $125/month gives heavy users much higher limits. Team is the plan most Australian small businesses should start on.

Is Claude AI free to use?

Yes, Claude AI has a free tier available at claude.ai. It includes access to Claude Sonnet and Haiku with limited daily messages. The free plan is suitable for occasional use and evaluation, but business users will likely need the Pro or Team plan for reliable, uninterrupted access.

What is the difference between Claude Pro and Enterprise?

Claude Pro is $20 per month for individual users and offers higher message limits plus priority access. Enterprise adds unlimited usage, admin controls, SSO integration, custom data retention policies, usage analytics, and dedicated support with SLAs. Enterprise pricing is custom and based on team size.

Is Claude AI cheaper than ChatGPT?

Claude and ChatGPT are priced similarly for consumer plans — both offer free tiers and $20 per month Pro subscriptions. At the API level, Claude Sonnet is competitively priced against GPT-4o. The real cost difference comes down to which model performs better for your specific use case, as a cheaper model that requires more retries ends up costing more.

Can I use Claude AI for my business without the API?

Yes. Many businesses use Claude effectively through the Pro or Enterprise chat interface without touching the API. The chat interface handles document analysis, writing, research, and ad-hoc tasks well. You only need the API if you are building Claude into your own software products or automating high-volume workflows.

Build With Claude — the Smart Way

Choosing the right Claude plan is step one. Getting real value from it is the harder part.

At Sandlabs, we are an AI development studio that builds custom solutions powered by Claude and Anthropic's API. We help businesses move from "we have AI subscriptions" to "AI is saving us 20 hours a week."

What we do:

• Custom AI agents and automations built on Claude
• API integrations that connect Claude to your existing tools
• Cost optimisation for teams already using Claude at scale
• Fixed-price projects from $5K to $60K, delivered in 2-6 weeks

Whether you need help choosing the right plan, optimising your API spend, or building something entirely new — let's talk.

Working in Australia? We run engagements out of Sydney, Melbourne, Brisbane, and Toowoomba on AU hours with AU data residency. If you already know you want the work done rather than advised on, our AI automation agency page covers how we build and ship automations end to end. For broader context on what an AI engagement looks like, read our AI consulting guide.