50% off every model
Sonnet 4.6 at $3/M input and $15/M output. Opus 4.7 at $5/M input and $25/M output. Haiku 4.5 at $1/M input and $5/M output. All half the official rate.
There are four ways to pay for Claude Code: the free tier, Pro at $20/month, Max at $100/month, or a pay-per-token API key. RunAPI mirrors every Claude model at half the official rate for the API route.
Claude Code itself is free to install. What you pay for is model usage. You can use a subscription with fixed monthly limits, or an API key billed per token. The cards below summarize each route before the full comparison.
Limited Claude Code usage at no cost. Enough to try the tool, not enough for daily work.
Subscription with Sonnet access and modest daily limits. Best for light, individual use.
Higher limits with Opus access. The $200 tier adds 5x the usage for heavy users.
Pay per token with no cap. Sonnet 4.6 at $3/M input and $15/M output — 50% off official.
The subscriptions trade a fixed monthly fee for usage caps. The API trades per-token billing for no cap and full model choice. The table compares all four so you can match the plan to your workload.
| Plan | Price | Models & limits | Best for |
|---|---|---|---|
| Free | $0 | Limited Sonnet, low daily cap | Trying Claude Code |
| Pro | $20/mo | Sonnet, modest daily limits | Light individual coding |
| Max | $100/mo ($200 for 5x) | Sonnet + Opus, high limits | Heavy daily users |
| API via RunAPI | Pay-per-token | All Claude models, no cap | Teams, CI, multi-model setups |
Subscriptions cap usage by message or session over a time window. The API has no message cap — you pay for the tokens you use and scale freely.
The cheapest route depends on how much you code. The table estimates the monthly API cost through RunAPI at four usage levels and shows when a subscription becomes the better deal.
| Usage level | Approx output tokens/mo | API cost via RunAPI (Sonnet 4.6) | Cheaper route |
|---|---|---|---|
| Occasional | ~1M | ~$15/mo | API or Pro |
| Daily light | ~3M | ~$45/mo | API |
| Daily heavy | ~6.7M | ~$100/mo | Break-even with Max |
| Power user | ~15M+ | ~$225/mo+ | Max ($100 or $200) |
Estimates use Sonnet 4.6 at $15/M output through RunAPI. Input tokens and cache discounts lower the real total. The break-even with Max ($100/mo) lands around 6.7M output tokens per month.
Instead of a fixed subscription, you point Claude Code at the RunAPI endpoint and pay per token at 50% of Anthropic's published rates. There is no cap and every Claude model is available.
Sonnet 4.6 at $3/M input and $15/M output. Opus 4.7 at $5/M input and $25/M output. Haiku 4.5 at $1/M input and $5/M output. All half the official rate.
Unlike Pro and Max, the API does not throttle you after a daily limit. You pay for what you use and keep working through long sessions.
Claude Code reuses the same system prompt and file context on every request. Cache reads cost 90% less, so real-world bills run below the headline per-token rate.
Set CLAUDE_CODE_MAX_OUTPUT_TOKENS to cap tokens per response. Lower caps cut cost on verbose tasks; raise it when you need long edits in one pass.
Sign up at runapi.ai and create an API key from Dashboard → API Keys. No credit card required for the free tier.
Export ANTHROPIC_BASE_URL=https://api.runapi.ai so Claude Code routes requests through RunAPI instead of the default endpoint.
Export ANTHROPIC_API_KEY with your RunAPI key. Claude Code uses these two variables on startup — no other config change is needed.
Launch claude in your project. Pick claude-sonnet-4-6 or claude-opus-4-7 and code at 50% of the official rate, with usage visible in the RunAPI dashboard.
Claude Code is free to install and there is a free usage tier with a low daily cap. To use it for real daily work you need either a Pro or Max subscription, or an API key. The API route through RunAPI is pay-per-token at 50% off.
It depends on the route. Pro is $20/month, Max is $100/month ($200 for the 5x plan). The API route has no fixed fee — at light daily use it runs around $45/month through RunAPI, and around $100/month at the Max break-even point.
For most individual developers, yes. The break-even on Sonnet 4.6 through RunAPI is around 6.7M output tokens per month, the point where API cost reaches the $100 Max fee. Below that the API is cheaper; above it, Max wins on a flat fee.
The API cost is per token. You pay for input and output tokens at the model's rate. Through RunAPI, Sonnet 4.6 is $3/M input and $15/M output — half the official Anthropic rate. Cache reads cut the input cost by a further 90%.
Yes. Set ANTHROPIC_BASE_URL to https://api.runapi.ai and ANTHROPIC_API_KEY to your RunAPI key. Claude Code reads both on startup and routes every request through RunAPI without any other change. Persist the two exports in your shell profile so they survive new sessions.
It caps how many tokens Claude Code can return in a single response. Lowering it reduces cost on verbose tasks and prevents runaway output. Raise it when you need long edits or large file rewrites completed in one pass. Set it as an environment variable before launching.
The Claude Code free tier covers limited in-app usage, not programmatic API access. For API access you create a RunAPI account, which has its own free credits to test any Claude model before you fund a pay-as-you-go balance. The two free tiers are separate and do not overlap.
Teams usually pick the API route. It has no per-seat subscription, no message cap, works in CI pipelines, and gives every model from Haiku to Opus. Through RunAPI the whole team pays one pay-as-you-go balance at 50% off official rates.
Create a free RunAPI account, set two environment variables, and code in Claude Code with Sonnet, Opus, or Haiku at 50% off official Anthropic pricing.