Free Claude plan calculator · No signup

Pro, Max, or API — which pays off?

Tell it how you actually use Claude Code and get a straight answer: the cheapest plan that fits, what your usage is worth at API prices, and when you'd hit each plan's limits.

Model mix
How you use it

≈ 14h/week (14h Sonnet) · API-equivalent ~$480/month

◆Pro fits your usage and costs $20/mo vs ~$480/mo at API prices — you'd be getting ~24.0× the value.
  • $Pay-as-you-go API — ~$480/moYour usage priced at API rates. Every subscription below that number is a discount.
  • ✓Pro — $20/mo · RECOMMENDEDFits comfortably. Your ~14h Sonnet/week sits under the expected floor (40h). Comfortable headroom. Sonnet in Claude Code. No meaningful Opus access.
  • ✓Max 5x — $100/mo Fits comfortably. Your ~14h Sonnet/week sits under the expected floor (140h + 15h Opus). Comfortable headroom. ≈5× Pro limits, real Opus access.
  • ✓Max 20x — $200/mo Fits comfortably. Your ~14h Sonnet/week sits under the expected floor (240h + 24h Opus). Comfortable headroom. ≈20× Pro limits — built for all-day agents.

Estimates only — limits are dynamic and vary with codebase size, caching, and demand. Prices as of July 2026.

Free · instant · based on published prices and Anthropic's expected-usage ranges

The plans

Claude Pro vs Max 5x vs Max 20x vs API

All four buy the same models — what changes is how much usage is bundled, and whether Opus is realistically available.

🟢

Pro — $20/mo

Everyday tier: Claude apps + Claude Code with ~40–80 Sonnet hours/week expected. The default for most developers; no meaningful Opus.

🔵

Max 5x — $100/mo

≈5× Pro's limits plus real Opus access (~15–35 h/week). The sweet spot for daily multi-hour Claude Code work.

🟣

Max 20x — $200/mo

≈20× Pro. For people running Claude Code (and subagents) most of the day, every day. Heavy users extract thousands of $ in API-equivalent value.

⚙️

API — pay-as-you-go

No caps, no monthly fee — you pay per token. Cheapest for light/bursty use, most expensive for daily grinding. Also what teams use for CI and automation.

How limits work

5-hour windows, weekly caps, dynamic ceilings

▹

The 5-hour session window

Usage is tracked in rolling 5-hour windows that start with your first message. Hit the window cap and you wait for it to reset — this is the limit casual users feel.

▹

The weekly cap

On top of session windows there's an overall weekly ceiling (resets every 7 days). This is the one heavy Claude Code users hit — it's why the calculator reports usage per week.

▹

Why 'expected ranges', not numbers

Anthropic publishes ranges (like 40–80 Sonnet hours/week on Pro) because actual consumption depends on codebase size, conversation length, caching, and current demand. Big monorepo + long sessions = the low end.

▹

Opus burns ~5× faster

Opus tokens cost about five times Sonnet's, and plans meter them accordingly. If you default to Opus, expect your effective hours to shrink dramatically — the mix selector above models exactly this.

FAQ

Claude plan questions

Is Claude Max worth it?+
If you use Claude Code more than ~2 hours a day, almost certainly. At moderate intensity that's ~$300+/month of API-equivalent usage — 3× the price of Max 5x. Run your numbers above; the multiple is usually the convincing part.
Can I use Claude Code on the Pro plan?+
Yes — Pro includes Claude Code with Sonnet. What Pro doesn't give you is meaningful Opus access or the headroom for long agentic sessions; that's what the Max tiers add.
What happens when I hit my limit?+
Claude tells you which window you've exhausted and when it resets (session windows reset within 5 hours; the weekly cap on a 7-day cycle). On Max plans you can also fall back to API billing for the gap if you've connected a key.
Should I just use an API key instead?+
Only if your usage is light, bursty, or automated (CI pipelines, scripts). For steady interactive coding, subscriptions bundle tokens at a steep effective discount — the calculator shows the exact multiple for your pattern.
Do these numbers include Claude.ai chat usage?+
The plan limits are shared across Claude.ai, the apps, and Claude Code. The calculator models Claude Code hours since that's what dominates most developers' usage; add a bit of headroom if you also chat heavily.
How do I see my real usage?+
Run npx ccusage@latest for API-equivalent spend from Claude Code's local logs, or /usage inside Claude Code for your live plan meters. Then feed the reality back into this calculator — and turn it into a shareable card with our Claude Code Wrapped tool.
Go further

Whatever plan you pick, ship faster on it

The ClaudeThings kits squeeze more out of every session window: production-grade agents, skills, and slash commands for engineering and marketing work.

🛠️
Most popular

Engineering Kit

Battle-tested Claude skills & workflows for builders.

  • ✓ 58 agents · 61 skills · 159 commands
  • ✓ Security, testing & code-review skills
  • ✓ Every skill formatted to spec
  • ✓ Lifetime updates
Get the Engineering Kit
📣
For founders

Marketing Kit

Turn Claude into your growth team.

  • ✓ 31 agents · 42 skills · 32 commands
  • ✓ Content, SEO & email-sequence skills
  • ✓ Landing-page & funnel templates
  • ✓ Lifetime updates
Get the Marketing Kit

Or grab all 89 agents, 103 skills & 181 commands in the Complete Bundle →