Search "best startup accelerators" and you get the same 15 names in a different order. Y Combinator, Techstars, 500, repeat.
Those 15 accept a fraction of a percent of applicants. The other 258 programs in the US, the ones with a real acceptance rate and a check that clears, do not make the listicles because nobody bothers to compile them.
We compiled them.
This is a free, ungated database of 273 startup accelerators operating in the United States, spread across 46 states. Every row carries nine fields: program name, city, state, domain, what the program actually offers, admission criteria, direct application form URL, target segmentation, and application timeline.
Below we broke down what 273 programs tell you about money, equity, verticals, and geography. Use it to shortlist in an afternoon instead of a month.
What you will find in this article
- What is inside the US startup accelerator database, field by field
- Startup accelerator vs incubator vs venture studio, defined properly
- What 273 programs reveal about funding amounts, equity terms, and program length
- The geography of American accelerators, and why 46 states matters more than 2
- How to shortlist five programs that fit your stage, sector, and dilution tolerance
- Why B2B sales teams should treat this list as a distribution channel
- FAQ: equity, acceptance rates, international founders, and application timing
What is a startup accelerator?
A startup accelerator is a fixed-term program that gives early-stage companies capital, mentorship, and a network in exchange for equity or, in a growing number of cases, nothing at all. Programs typically run 8 to 12 weeks and end with a demo day in front of investors.
The category gets muddled constantly, so here is the clean version:
Accelerators are for companies that already exist and need to move faster. If you have a slide deck and a co-founder you met last week, you are looking for an incubator.
What 273 US accelerators tell you about the market
We ran the analysis across every row in the database. Six findings change how you should approach applications.
1. The median accelerator check is $100,000
Ninety-three programs disclose a specific investment amount. The distribution is wider than most founders expect.
The median disclosed check is $100,000. Y Combinator sits at the top of the range with $500,000 per company, structured as $125,000 for 7% on a post-money SAFE plus $375,000 on an uncapped SAFE with MFN.
The tail matters more than the headline. Thirty-two programs write checks under $25,000, which is not capital, it is a stipend. If a program takes 6% for $20,000, you are valuing your company at $333,000 in exchange for mentorship. Sometimes that trade is worth it. Usually it is not.
2. Thirty-five programs take no equity at all
Sixty-nine programs explicitly reference equity in their terms. Thirty-five state that they take none, running on non-dilutive or grant-funded models.
Where an equity percentage is stated, the median stake is 7%. That number is remarkably consistent, which tells you the market has standardized around the YC benchmark whether or not the program delivers YC-level value.
The equity-free segment is largely university programs, state economic development initiatives, and corporate-sponsored accelerators. They give less capital and less prestige. They also give you your cap table back.
3. Twelve weeks is the default, but a third of the market runs differently
Twelve weeks is the most common program length in the database, followed by 8 weeks and 3-month cohorts. Ten programs run 6 months or longer.
Format splits too. Forty-one programs run virtual or remote tracks. Thirty-one require in-person attendance or relocation. Forty-three include physical office space or coworking as part of the offer.
Relocation is the hidden cost nobody prices in. A 12-week in-person program in San Francisco costs a two-person founding team roughly $20,000 in housing and living expenses. If the check is $50,000, half of it goes to rent.
4. Healthcare and climate outnumber AI
Founders assume every program has pivoted to AI. The data says otherwise.
Vertical-focused programs beat generalist programs for most founders. The mentor bench actually understands your regulatory path, the demo day audience actually writes checks in your category, and the alumni network actually sells to your buyer.
Nine programs describe themselves as industry-agnostic. Those are the ones competing directly with YC, and they compete badly.
5. Mentorship is the real product, not the money
One hundred and eighty of 273 programs reference mentorship as a core component. Forty-one run a demo day. Forty-three provide space.
That is the honest hierarchy. Accelerators sell access, and capital is the wrapper. When you evaluate a program, evaluate the mentor list and the alumni outcomes, not the check size.
Ask one question in every conversation with a program alum: "Which specific introduction changed your trajectory?" If nobody can answer it, the network is decoration.
6. Geography is far less concentrated than venture capital
California holds 81 programs and New York holds 30. Together that is 111 of 273, which is real concentration but nothing like the funding map.
The list spans 46 states. At city level, New York leads with 21 programs, ahead of San Francisco with 14, Los Angeles with 10, Palo Alto with 9, and Atlanta with 8.
Atlanta with 8 programs and Nashville with 4 are not accidents. State economic development money funds accelerators in places venture capital ignores, which means acceptance rates are better and the local ecosystem is hungrier for wins.
Fifty-five programs reference university or student affiliation in their criteria. Twenty require you to be based in, or relocate to, a specific region. Both filters cut your realistic list faster than any ranking will.
How to shortlist five accelerators in one afternoon
Applying to 40 programs is a signal that you have not decided what you need. Here is a filter that gets you to a real shortlist.
Step 1: Decide what you are actually buying
Write down the one thing the program must deliver. Capital, customer introductions, technical mentorship, regulatory guidance, or credibility for the next round. One thing, not five.
Most founders skip this and end up in a program that gives them a Slack channel and a t-shirt.
Step 2: Filter by segmentation, then by stage
Use the target segmentation column first. If a program lists healthcare and you sell developer tools, you are not a borderline case, you are a no.
Then filter by stage. In the database, 88 programs specify early-stage, 32 mention seed, 24 mention pre-seed, and 28 accept idea or pre-revenue companies. Applying above your stage wastes a cycle. Applying below it wastes a year.
Step 3: Price the dilution against the check
Run the math before you fall in love with a logo.
If the implied valuation is below what you could raise from a single angel next quarter, the program has to deliver something the angel cannot. Sometimes it does. Make it prove that.
Step 4: Check the timeline before you write anything
Sixty-one programs run cohort or batch cycles with fixed deadlines. Thirty accept rolling applications. Forty state an explicit deadline.
Cohort programs punish late applications hard. Y Combinator, for example, sets an on-time deadline of February 9 with decisions by March 13 for the April to June batch. Miss it and you wait a quarter.
Be aware that 87 rows in the database have no timeline listed. For those, check the program site before you build a plan around them.
Step 5: Apply to five, seriously
Five well-researched applications beat 30 templated ones. Reference the program's actual thesis, name a mentor whose experience matches your problem, and answer the traction question with numbers.
One hundred and ninety-nine programs in the database include a direct application form URL, so the friction is already removed.
Why B2B sales teams should read this list differently
If you sell to startups, an accelerator is not a program. It is a pre-qualified cohort with a shared calendar, a shared Slack, and a partner who wants their portfolio to succeed.
One accelerator relationship gives you access to 20 to 100 companies at the same stage, with the same problems, at the same moment. That beats cold outreach on economics alone.
The play is straightforward. Filter the database to programs whose target segmentation matches your ICP. Approach the program director with something the cohort needs, usually a workshop, a tool credit, or a template pack. Deliver it without a pitch. Let the portfolio come to you.
For most B2B tools selling into early-stage companies, this is the cheapest pipeline available and the least crowded channel on the list.
Four mistakes founders make with accelerator lists
Mistake 1: Optimizing for prestige over fit. A vertical program with 12 relevant mentors will do more for you than a famous generalist program where you are the only company in your category.
Mistake 2: Ignoring the equity math. Six percent feels small in a spreadsheet. It is not small across three rounds of dilution, and an accelerator stake sits on your cap table forever.
Mistake 3: Applying to everything. Program directors read hundreds of applications and can spot a template instantly. Volume works in outbound email. It fails in accelerator applications.
Mistake 4: Treating demo day as the goal. Demo day is a 3-minute pitch to a room that has seen 40 pitches that morning. The introductions you build in week 4 are worth more than the stage in week 12.
FAQ: US startup accelerators
Is the database free? Yes. No gate, no form, no email required. Open the sheet, make a copy, and filter it however you need.
How many accelerators does it include? 273 programs across 46 US states, with 199 including a direct application form link.
Do all accelerators take equity? No. Thirty-five programs in the database explicitly operate without taking equity, typically university-affiliated, government-funded, or corporate-sponsored programs. Where equity is taken, the median stake is 7%.
How much money do accelerators actually invest? Among the 93 programs that disclose an amount, the median is $100,000. Thirty-one programs invest $250,000 or more, and 32 invest under $25,000.
Can international founders apply to US accelerators? Many accept international companies, though visa requirements and incorporation rules vary by program. Check the admission criteria column, and confirm on the program site before applying.
How long does an accelerator program last? Telve weeks is the most common length in the database, followed by 8-week and 3-month formats. Ten programs run 6 months or longer.
What is a realistic acceptance rate? Top national programs accept roughly 1% of applicants. Regional and vertical programs typically run far higher, which is exactly why the other 258 names on this list are worth your attention.
How often is the database updated? Periodically. Application deadlines shift, so always verify the timeline on the program's own site before you build a plan around a date.
What to take away
- There are 273 accelerators across 46 states. The famous 15 are a rounding error in your options.
- The median check is $100,000 and the median equity stake is 7%. Any program asking more than that has to justify it.
- Health and climate programs outnumber AI programs. Vertical fit beats brand name for almost every founder.
- Thirty-five programs take no equity. If dilution is your constraint, start there.
- If you sell B2B into startups, this list is a distribution map, not a directory.
Choosing a program is a founder's problem. Building the pipeline that makes the program worth it is ours.
If you want the outbound motion built around your ICP, whether that runs through accelerator cohorts or direct, book a free consultation.
.avif)
