Tokens for Good logo Tokens for Good

Does Claude Pro include API credits?

No. A Claude Pro subscription includes no API credits. The Claude subscription (Pro, Max, Team) and the Claude Developer Platform (the API) are separate products on separate billing: paying for Pro gives you nothing to spend on the API, and buying API credits does not raise your subscription usage limits. They are two ways to pay for the same models. You can hold both, but neither one funds the other.

Two products, two bills

Same company, same models, different products. That is why the question comes up constantly.

There is no credit transfer in either direction. A Max plan does not top up an API balance, and an API balance does not extend a five-hour window or a weekly cap.

What each product actually covers

The subscription covers you, the person. It gives you the Claude apps, Projects, connectors, and Claude Code, all metered against a rolling five-hour session window plus weekly caps that refill on their own. There is no API key involved, so you cannot point an application at it, and building a product on top of it is not what it is for. For the mechanics of those windows, see Claude weekly usage limits, explained.

The API covers your software. It gives you keys, per-token billing, prompt caching, the Batch API, and the operational controls a production system needs. It has no usage windows and nothing resets; it has a credit balance that draws down, or a monthly invoice for organizations set up that way.

Where the line blurs: Claude Code

Almost all of the confusion traces to one product. Claude Code runs on either billing model. Authenticate it with a Pro or Max subscription and it draws on your usage windows at no extra cost. Point it at an API key instead and every session is metered and lands on your API invoice.

One command tells you which one you are on. /usage (/cost is an alias) shows a session block with token counts and a dollar figure computed locally at list rates, plus, on a Pro or Max login, your plan usage bars: how much of each window is used and when it next resets. If you see plan usage bars, the dollar figure is informational and no per-token bill is accruing. If you see only the session cost, you are on API billing. One trap: if an ANTHROPIC_API_KEY environment variable is set, Claude Code uses it in preference to your subscription login, which is the most common way people land on API billing by accident.

One more line that moved in 2026. Anthropic announced, then paused on June 15, 2026, a separate monthly credit for programmatic use on paid plans (the Claude Agent SDK, claude -p, Claude Code GitHub Actions). As of August 2026 that credit is not live and programmatic use still draws on your subscription limits. Either way it is plan usage, not API credits.

Which one you actually want

Pick the subscription if a human is doing the work: writing, research, and interactive coding, daily and heavily. The flat fee caps your spend and removes per-token anxiety, and consistent heavy use gets more work per dollar out of a subscription than out of metered billing. Claude Max vs Pro covers where the tiers land.

Pick the API if software is calling Claude: a product feature, a pipeline, a scheduled batch job, anything that needs keys, logs, and reproducible billing. It also wins when your use is genuinely sporadic, since an idle day costs zero. The full comparison is in Claude Pro vs API: the real cost.

Plenty of people want both, and that is a reasonable setup: a subscription for hands-on work, an API key for anything running unattended in production. Just do not expect one bill to reduce the other.

The surplus a subscription leaves behind

Here is the asymmetry worth acting on. API credits sit there until you spend them. Subscription capacity does not: it is a ceiling, not a balance, and whatever you do not use before the window resets is gone. Most subscribers finish most weeks well under the cap, so real paid-for capacity quietly expires every cycle. Are you wasting your Claude subscription? walks through the signs.

Tokens for Good captures that surplus. Your Claude Code claims a queued nonprofit, researches its real-world impact against a fixed methodology with citations, and submits a structured report. Every organization is researched twice by independent contributors, validated, consolidated, scored deterministically, and human-reviewed before it reaches the public directory. It runs on the subscription you already pay for with no separate API cost, and it can run on a schedule in the background. "Tokens" here means AI model tokens, not crypto: no coin, no wallet, no blockchain. See how the research works.

Frequently asked questions

Does Claude Pro include API credits?
No. The subscription and the Claude Developer Platform are separate products with separate billing. A Pro or Max plan grants no API credits, and there is no way to convert subscription value into API spend.
Does Claude Pro include API access?
No. API access requires a separate Developer Platform account with its own payment method, paid for with prepaid usage credits per token. Your Pro login does not produce an API key.
Do API credits increase my Claude Pro usage limits?
No. API credits and subscription limits are unrelated. Your five-hour session window and weekly caps are set by your plan, and buying API credits does not raise them. If you need more subscription headroom, you upgrade the plan or buy usage credits for the plan itself.
Does Claude Code use my subscription or the API?
Either, depending on how you authenticate it. On a Pro or Max login it draws on your subscription usage windows at no extra cost; on an API key it is metered per token. Run /usage: on a subscription login it shows plan usage bars, and its session block shows a locally computed dollar figure that only matters on API-key billing.
Anthropic subscription vs API: which is cheaper?
For heavy, consistent hands-on use a subscription usually delivers more work per dollar, because the fee is flat while your usage rises toward the ceiling. For sporadic use, or for software calling Claude, metered API billing is often cheaper since idle time costs nothing.

No API credits, but plenty of surplus

Subscription capacity you never reach expires at every reset. Tokens for Good turns it into verified nonprofit research at no separate API cost.

See how Tokens for Good works