Claude for Nonprofits (2026): The $8 Team Plan Seat, Who Qualifies, and What a 10-Person Team Actually Pays

Claude for Nonprofits: the short answer

Somebody on your board read that AI is "free for nonprofits." It isn't, but it is close. As of September 2026, Anthropic lists the Claude Team plan at $8 per user per month for verified nonprofits, against $20 per seat billed annually or $25 monthly at the standard rate. Nonprofits in low- and middle-income countries are listed at $3. The minimum is two seats. If you have fewer than 20 people, you verify through Goodstack in a few minutes and the discount is applied when you sign up; 20 or more, Anthropic sends you to sales. Below: who qualifies, what the $8 seat includes, what it doesn't say, and what a real 10-person team pays, with every number read off Anthropic's nonprofit and pricing pages on September 28, 2026. The cross-vendor version, ChatGPT and Google included, is AI for nonprofits.

What Claude for Nonprofits costs

Against the annual standard seat, $8 is 60% off; against the monthly seat, it is 68% off. The $3 rate is 85% off the annual price. The standard Team plan and what it includes are on Claude Team plan.

What the page does not say is just as useful. It does not say whether the $8 rate is billed annually or monthly, it does not price the Premium seat for nonprofits, and it does not mention an API discount. If any of those decide your purchase, ask before you sign up rather than assuming the standard plan's rules carry over.

Who qualifies

Anthropic's eligibility list is broader than most people expect:

• Registered 501(c)(3) organizations and their international equivalents
• K-12 public and private schools
• Certain mission-based healthcare providers: independent Critical Access Hospitals, Rural Emergency Hospitals, Federally Qualified Health Centers, and Rural Health Clinics

That last line matters. A rural health clinic is exactly the kind of organization that assumes enterprise AI is out of its budget, and it is on the list.

How to get it: the 20-person line

The sign-up path splits on headcount. Fewer than 20 people: you verify through Goodstack, which Anthropic says takes about two to three minutes, and once you are approved you get an email to finish setup with the discount already applied. 20 or more: you talk to Anthropic's nonprofit sales team, and Enterprise customers are verified during that sales process.

Goodstack is the same verification partner OpenAI uses for its nonprofit program, so if you have already been through it for ChatGPT, you know the drill. The ChatGPT side of the comparison is on ChatGPT for nonprofits.

What the $8 seat includes

• Claude and Claude Code. The same assistant and coding tool the standard Team plan gets.
• Nonprofit connectors. Anthropic names Blackbaud, Benevity, and Candid: your donor database, giving platform, and foundation research, readable from inside Claude.
• File creation. Claude produces docs, slides, spreadsheets, and PDFs directly.
• Admin basics. Single sign-on, domain capture, and central billing.
• Training. A free "AI Fluency for nonprofits" course aimed at everyone from executive directors to program coordinators.

Enterprise adds more usage, a bigger context window, a Compliance API, SCIM, and custom data retention. Anthropic also says its plans are configurable to meet SOC 2 Type II, ISO 27001, ISO 42001, and CSA Star compliance needs, which is the sentence to forward to the board member who asks about donor data.

The math on a 10-person team

• Nonprofit Team: 10 seats at $8 is $80 a month, $960 a year.
• Standard Team, annual: 10 at $20 is $200 a month, $2,400 a year.
• Standard Team, monthly: 10 at $25 is $250 a month, $3,000 a year.
• Nonprofit access rate: 10 at $3 is $30 a month, $360 a year.

The verification is a few minutes of work for $1,440 to $2,040 a year back. The expensive mistake is not the subscription at all. It is buying ten seats, sending a "we have AI now" email, and discovering in March that three people use it. Pick the two workflows that eat the most staff hours, usually grant drafts and board or donor reporting, and set those up before you roll it out to everyone.

Which model should a nonprofit use?

The seat price does not change the model decision. Sonnet is the everyday default for drafting and summarizing. The long jobs, a full grant application built from a program budget and outcome data, or a year of donor records turned into a board report, are where Anthropic's new Opus 5.5 earns its place, at a lower price than the Opus it replaced. What changed with it is on Claude Opus 5.5.

What to do today

If you have fewer than 20 people, go through the Goodstack verification before your next staff meeting; it takes less time than the meeting's first agenda item. Then write down the one document your team rebuilds every month (the grant report, the board packet, the donor update) and make that the first thing Claude does for you. If you would rather have it set up properly, with your donor system connected and the first workflows built, that is what the Claude install covers, from $4,500 in about two weeks, with your seats paid to Anthropic at the nonprofit rate.