Every founder who has ever filled out the YC application has Googled the same thing at 1 a.m.: what is the acceptance rate? The answer most people land on — '1%' — is technically close, but it hides almost everything that actually matters. The real picture is more nuanced, more encouraging in some cases, and more brutal in others.
Y Combinator does not publish a clean acceptance-rate dashboard. The numbers below are pieced together from official YC blog posts, partner interviews on the YC podcast, Garry Tan's letters, and Startup School lectures. Where a figure is an estimate, it's labelled.
The headline number
YC has stated multiple times — most recently in the official YC blog and in talks by Michael Seibel — that they typically receive somewhere between 15,000 and 25,000 applications per batch and accept roughly 200–250 companies. That puts the raw acceptance rate at roughly 1.0% to 1.5% per batch.
But 'applications' includes a long tail of one-line submissions, missing co-founders, and ideas in industries YC has publicly said it does not fund. If you filter for completed, two-founder, software-shaped applications, founders who have been accepted estimate the realistic rate is closer to 3–5%.
Acceptance has trended down as YC has scaled up
| Year | Batch | Companies funded | Notes |
|---|---|---|---|
| 2005 | Summer | 8 | First-ever batch (Reddit, Twitch precursor Justin.tv era follows) |
| 2010 | Summer | 36 | Airbnb is a YC alum at this point |
| 2014 | Winter | 85 | Sam Altman becomes president |
| 2016 | Summer | 120 | First time over 100 companies |
| 2018 | Summer | 132 | |
| 2021 | Winter | 414 | All-time high — fully remote pandemic batch |
| 2023 | Winter | 282 | Garry Tan returns YC to in-person, smaller batches |
| 2024 | Summer | 256 | |
| 2025 | Spring | 240 | YC moves to 4 batches per year |
Where your odds are quietly higher
- RFS-aligned ideas. When YC publishes a Request for Startups, founders building in those areas anecdotally see meaningfully higher interview rates. Partners openly admit to filtering for RFS themes during reading week.
- Technical founders building hard things. Hardware, biotech, defense and crypto founders compete in much smaller pools than the consumer-AI flood.
- Repeat founders. YC funds second-time founders at a higher rate, even when the first company failed.
- Founders with a working demo. A 60-second Loom of a working product moves applications from 'maybe' to 'interview' faster than any other single signal.
What does NOT move the needle
- Going to a famous school. YC reads applications without filtering on school.
- Having a perfectly polished deck. There is no deck in the YC application.
- Being in San Francisco already. YC will pay to move you.
- Long answers. Partners read thousands of apps; the highest-signal answers are the shortest.
Key takeaways
- Headline acceptance rate is ~1–1.5%. Realistic rate for serious applicants is 3–5%.
- Batch sizes peaked at 414 in W21 and have since shrunk back to 240–280.
- Building inside a current RFS theme is the single highest-leverage edge.
- A working demo Loom outperforms every other application asset.