Short answer. Partly. As of November 17, 2026, Anthropic's pricing page says "New users receive free credits to test the API," but it does not publish the amount, and after that the Claude API runs on prepaid credits you buy in the Console. The only free dollar credit with a published size comes with a paid chat plan: Max includes $100 a month of API credit, Ultra $200, and Team $20 per Standard seat or $100 per Premium seat, pooled and capped at $500.

This guide is for founders, developers and small teams who searched for a free Claude API key. It covers what a new account gets, every free path we could find on Anthropic's own pages and who qualifies, what $20 or $100 of credit buys on each current model, and the billing settings to check before the first call. We read every source on 17 November 2026 and link each one where it is first cited; the request counts are our arithmetic.
Key numbers
| Item | Number |
|---|---|
| New-user test credits (pricing FAQ) | Not published |
| Max plan API credit (a month, after 7 days on the plan) | $100 |
| Ultra plan API credit (a month) | $200 |
| Team plan API credit, Standard seat (per seat, pooled) | $20 |
| Team plan API credit, Premium seat (per seat, pooled) | $100 |
| Team pool cap (a month) | $500 |
| Free code execution (a month, then $0.05 an hour per container) | 1,550 hours |
| Start tier monthly spend cap (Build $1,000, Scale $200,000) | $500 |
| Purchased credit expiry (non-refundable) | one year |
| Batch API discount (input and output tokens) | 50 percent |
Read on 17 November 2026 from Anthropic's pricing, rate limits and token counting docs, three Claude Help Center articles and the Supplemental Credit Terms.
What a new Claude Console account gets
Anthropic's help article for new organizations says that to use the API and the playground you "add usage credits to your organization's balance," fill in details about your organization and use case, then "enter your payment details and fund your account" (Claude Help Center).
The billing article is plain about the model: "Claude API and playground usage is billed through prepaid usage credits. Buy credits before you use the API." When the balance is gone, "you can no longer call the API or use the playground until you add more."
The starter credit is real, but its size is not published. The pricing page FAQ answers "Are there free tiers or trials" with "New users receive free credits to test the API. Email sales for information about extended trials for enterprise evaluation." Anthropic's Supplemental Credit Terms (effective March 4, 2024) say that free credits Anthropic offers "as part of a promotion or during sign up" are Promotional Credits, which "expire at the time indicated when issued (or otherwise one calendar year from when issued if no time is specified)." We found no Anthropic page that states the amount, so treat it as a test balance, not a plan.
Neither billing article we read states a minimum credit purchase.
Every free path we found, and who qualifies
| Path | Who qualifies | Amount | Main limit |
|---|---|---|---|
| New-user test credits | New users | Not published | Promotional; expiry set when issued |
| Max plan API credit | Max subscribers, after 7 days | $100 a month | No rollover; one linked organization |
| Ultra plan API credit | Ultra subscribers, after 7 days | $200 a month | No rollover; one linked organization |
| Team plan API credit | Team Owner or Primary Owner claims, after 7 days | $20 per Standard seat, $100 per Premium seat | Pooled, capped at $500 a month |
| Token counting | Any organization | Free | Requests-per-minute limits by tier |
| Code execution hours | Any organization | 1,550 hours a month | Then $0.05 an hour per container |
| Extended trial for enterprise evaluation | Enterprises; email sales | Not published | Not published |
The Claude chat Free plan is not on this list. Anthropic's monthly credits article says "Free, Pro, and Enterprise plans aren't eligible" for API credit, Separately, Anthropic's consumer pricing page lists the chat Free plan at "$0" with "Limited usage every 5 hours and every week", a limit for the Claude apps, not an API credit.
Monthly API credits on Max, Ultra and Team
These are the largest published free credits, and on our reading they suit a builder who already pays for Claude. The help article sets the amounts: Max $100, Ultra $200, Team $20 per Standard seat and $100 per Premium seat, with Team seats "pooled into one monthly balance, capped at $500." Its own example: three Standard and two Premium seats receive $260 a month.
- Eligibility: "New subscribers can claim once they've been on an eligible plan for seven days," and "You don't need to add a card on Claude Platform to claim or use the credit."
- What it covers: the Messages and Message Batches APIs, the Console Playground, Claude Managed Agents and the Claude Agent SDK.
- What it does not cover: interactive Claude Code, extra usage in Claude, Claude Code or Claude Cowork, and Claude on Amazon Bedrock, Google Cloud Vertex AI or Microsoft Foundry.
- Expiry: "Unused credits expire at the end of each billing cycle," and the plan credit is spent before any credit you bought.
- When it runs out: with no other credit, "API requests stop until your next monthly credits arrive. Usage is never charged to your Claude plan."
Anthropic's claim steps are: on claude.ai open Settings, then Billing (Max or Ultra) or Organization settings, then Billing (Team Owner or Primary Owner); in the API credits section select "Link organization"; choose an existing Console organization or create one; then review the Supplemental Credit Terms and link it. You need the Owner, Admin or Billing role in that Console organization, and Anthropic suggests signing in to the Console with the same email you use on claude.ai.
We cover the claim flow and the Team pool in more depth in our guide to Claude Max and Team API credits.

What $20 or $100 buys on each Claude model
To make the credit concrete, we priced one example request: 1,000 input tokens and 500 output tokens, at the standard prices on Anthropic's pricing page for prompts under 100,000 tokens. Your requests will differ, so use the method rather than the totals.
| Model | Input / output per million tokens | Cost per request | Requests for $20 | Requests for $100 |
|---|---|---|---|---|
| Claude Haiku 5.5 | $0.10 / $0.50 | $0.00035 | 57,142 | 285,714 |
| Claude Sonnet 5.5 | $2 / $10 | $0.007 | 2,857 | 14,285 |
| Claude Opus 5.5 | $4 / $20 | $0.014 | 1,428 | 7,142 |
| Claude Fable 5.2 | $8 / $40 | $0.028 | 714 | 3,571 |
One lever stretches a small credit, and one price step shrinks it. The Batch API gives "a 50 percent discount on both input and output tokens" for work that can wait, which doubles the Haiku 5.5 count for $20 to about 114,000 requests. Haiku 5.5 also has a second price above 100,000 tokens of prompt ($0.50 input and $2.50 output per million), so long documents cost five times as much per token on that model.
Evaluation tier and spend caps for new organizations
A new organization may start with lower limits than the published tables. Anthropic's rate limits page says "New organizations and organizations with limited usage history may start in the Evaluation tier, with limits below the standard limits shown on this page," and that these limits "increase automatically as your organization builds usage history." It does not publish the Evaluation tier's numbers.
Each standard tier also carries a monthly spend cap: $500 on Start, $1,000 on Build and $200,000 on Scale. At the cap, "API usage pauses until 00:00 UTC on the first day of the next month" unless you request a higher limit. Our guide to Claude API rate limits has the per-model tables.
Free tools inside the paid API
- Token counting: "Token counting is free to use but subject to requests per minute rate limits based on your usage tier" (token counting docs). Count a prompt before you send it to see how many tokens it uses.
- Code execution: "Each organization receives 1,550 free hours of usage per month," then $0.05 an hour per container, and execution time has a minimum of 5 minutes. Code execution is free outright when the request includes the web search or web fetch tool versions dated 20260209 or later.
- Web fetch: "available on the Claude API at no additional cost"; you pay only for the fetched tokens that enter the conversation.
- Failed calls: "Failed requests aren't charged," though a request that "was on track to succeed" is still charged if your client disconnects or times out mid-way.
Billing settings to check before the first call
- Auto-reload. It buys credit when the balance reaches a threshold you set. If the balance is already at or below it, turning it on charges you right away.
- Your own spend limit. Anthropic says you can "set your own spend limit below your tier's cap"; when usage reaches it, requests return HTTP 400 with error type invalid_request_error until you raise or remove it.
- Expiry. "Credits expire one year from the purchase date, and the expiration date can't be extended."
- Refunds. "All credit purchases are non-refundable," so buy in small steps while you test.
- Which organization gets plan credits. You can link only one Console organization to a Max, Ultra or Team plan, and Anthropic says "You can't change the linked organization yourself" (support can).
Comparing providers for a first build? Our OpenAI API free tier guide covers the OpenAI side, and Claude for Startups covers Anthropic's startup offer. For what a deployed assistant costs per month once the free credit is gone, see AI chatbot for business cost.
How imisofts can help
We build AI agents and automations for small teams, including AI phone agents for home service businesses. If you want a build scoped against your real request volume, see our AI automation agency page.
Frequently Asked Questions
Partly. Anthropic says new users receive free credits to test the API but does not publish the amount, and ongoing free API credit comes only with a paid Max, Ultra or Team plan. Other usage is billed from prepaid credits you buy in the Claude Console.
Anthropic does not publish a figure. Its pricing FAQ says only that new users receive free credits to test the API. Its credit terms say free credits offered during sign up are Promotional Credits that expire when stated, or after one year if no time is given.
No. Anthropic's help article says Free, Pro and Enterprise plans are not eligible. Max includes $100 a month, Ultra $200, and Team $20 per Standard seat or $100 per Premium seat, capped at $500 a month.
Yes, if you claim monthly API credits from a Max, Ultra or Team plan. Anthropic says you do not need to add a card on Claude Platform to claim or use that credit. Otherwise you fund the account with prepaid credits.
API calls and the playground stop until you add credit. With plan credits and no other balance, requests stop until the next monthly credit arrives, and Anthropic says usage is never charged to your Claude plan.
We suggest Claude Haiku 5.5 and the Batch API where you can wait. At $0.10 input and $0.50 output per million tokens, $20 covers about 57,000 requests of 1,000 input and 500 output tokens, or about 114,000 with the 50% batch discount.