The Biggest Venture-Backed Exits by Year (2015–2026): IPOs, Acquisitions and What LPs Actually Got Back

The biggest venture-backed exits by year are the scoreboard that matters to limited partners, because only exits turn paper valuations into cash, and this article lists them from 2015 to October 2026: the largest IPOs, direct listings and acquisitions each year, the total exit value behind them, and what fund-level distribution data (DPI) say LPs actually received. It belongs to the Sectors & Exits by Year pillar of the Kurums Venture Capital hub, and reads alongside our analysis of how exits happen (IPO, M&A, secondaries) and the three sector reviews, starting with where venture capital went in 2015–2020.

What were the biggest venture-backed exits each year from 2015 to 2026?

The table below lists the largest venture-backed IPOs and acquisitions for each year from 2015 to October 2026, with the exit values at pricing or announcement and the US exit value recorded by the PitchBook-NVCA Venture Monitor for that year.

Values are approximate, at IPO pricing or deal announcement, from company filings, Crunchbase News, PitchBook-NVCA Venture Monitor (US exit value, Q4 2024 restated series; 2025 and 2026 from PitchBook reports) and press coverage. Some listed companies (Mobileye, Arm, Juniper, Splunk, Nuance, CXMT) were public or corporate-owned rather than classic VC-backed exits and are included for scale only.

Why was 2021 the record year for venture exits?

2021 produced $841.5 billion of US venture exit value (PitchBook-NVCA), nearly 2.5 times the 2020 record, because the IPO window, direct listings, SPAC mergers and strategic M&A all peaked in the same twelve months at peak public-market multiples.

The year’s listings read like a decade of backlog clearing at once. Coinbase’s April direct listing opened at a valuation near $86 billion; Rivian’s November IPO valued a company with almost no revenue at about $77 billion; Robinhood, UiPath, Roblox, Affirm, Bumble, Nubank, Didi, Toast, Freshworks, GitLab, HashiCorp and Samsara all listed; and SPAC mergers took Lucid, SoFi, Grab and dozens of others public. On the M&A side, Microsoft’s $19.7 billion purchase of Nuance was the largest pure venture-backed acquisition of the year, and Salesforce completed its $27.7 billion Slack deal. SPAC Research counted more than 600 SPAC IPOs raising over $160 billion in 2021, most of which were hunting for venture-backed targets.

Two caveats matter for anyone using 2021 as a benchmark. First, exit value is measured at listing or announcement, and a large share of 2021’s public cohort fell 50–90% within eighteen months, so the cash LPs actually realised depended on when their fund distributed or sold shares. Second, lock-ups and in-kind distributions meant much of the 2021 “exit value” reached LPs as volatile stock in 2022, which is one reason the 2021 exit record did not translate into the DPI one would expect. The sector composition of that year is covered in our 2021–2023 review.

How bad was the 2022–2024 exit drought?

US venture exit value fell from $841.5 billion in 2021 to $148.0 billion in 2022, about $120 billion in 2023 and $149.2 billion in 2024, with the IPO window effectively closed for 20 months and the largest 2022 deal, Adobe’s $20 billion offer for Figma, abandoned on regulatory grounds.

2022 saw almost no venture-backed technology IPOs of size in the US. The exits that did happen were smaller strategic acquisitions, take-privates of former venture darlings by private-equity firms (Zendesk, Anaplan, Coupa) and a handful of biotech listings. Adobe’s agreement to buy Figma for $20 billion in September 2022 would have been the largest venture-backed software acquisition ever; it was terminated in December 2023 after UK and EU regulators objected, with Adobe paying a $1 billion break fee, and Figma eventually listed in July 2025 at about $19.3 billion. The episode shaped the next three years of M&A: large acquirers avoided transformational deals and instead structured licensing-plus-hiring arrangements (Microsoft–Inflection for about $650 million in 2024, Google–Character.AI for about $2.7 billion, Amazon–Adept) that moved talent without a formal acquisition.

2023 tested the window in September with Arm (owned by SoftBank rather than venture funds, raising about $4.9 billion at a $54.5 billion valuation), Instacart (about $9.9 billion, far below its $39 billion 2021 mark) and Klaviyo (about $9.2 billion). All three traded poorly for months, and the window closed again. 2024 improved at the margin: Reddit ($6.4 billion), Rubrik, Astera Labs, Tempus AI, Ibotta and, in December, ServiceTitan listed; Pony.ai and Pinterest-style consumer names were scarce; and PitchBook-NVCA counted 1,259 US exits worth $149.2 billion, with Tempus AI, Astera Labs and Pony.ai among the six largest. Asia was the exception, with exit value above $143 billion in 2023, more than the US and Europe combined, mostly on Chinese and Indian listings.

Which 2025 exits reopened the market?

2025 reopened both routes: Google’s $32 billion acquisition of Wiz and Meta’s $14.3 billion investment in Scale AI led M&A, while CoreWeave, Circle, Chime, Figma, Klarna, Netskope, StubHub and Figure led a 40-IPO unicorn cohort worth about $207 billion at listing (Crunchbase); global exit value reached $549.2 billion (PitchBook).

Wiz was the defining deal. Founded in 2020, the cloud-security company had rejected a $23 billion Google offer in 2024 and accepted $32 billion in March 2025, the largest acquisition of a venture-backed company at the time and, according to Crunchbase, the main reason 2025 unicorn M&A reached about $67 billion across 36 deals, nearly nine times 2024’s roughly $7 billion. Meta’s purchase of a 49% stake in Scale AI for $14.3 billion in June, Naver Financial’s $10.3 billion acquisition of crypto exchange Dunamu, Palo Alto Networks’ $3.4 billion for Chronosphere, Marvell’s $3.3 billion for Celestial AI and ServiceNow’s $2.9 billion for Moveworks rounded out the top of the list. Cybersecurity and AI infrastructure dominated; Google also bought Windsurf’s leadership and licensing for about $2.4 billion in July.

The IPO cohort was the broadest since 2021. CoreWeave priced in March at about $23 billion and multiplied within months as AI-compute demand repriced; Circle priced in June at roughly $6.9 billion and tripled on stablecoin legislation; Chime listed at about $11.6 billion; Figma priced at about $19.3 billion in July and more than doubled on day one; Klarna listed in September at about $15 billion, well below its 2021 mark but above its 2022 down round; Netskope, StubHub, Figure, Via and Gemini followed. Crunchbase counts 40 traditional unicorn IPOs plus two SPAC or reverse-merger listings in 2025, with a combined listing value near $207 billion, in line with 2019 ($215 billion) and 2018 ($197 billion) though still about $90 billion below 2020. In the US, PitchBook-NVCA recorded 1,636 exits worth $297.8 billion, up 92.7% on 2024.

What the 2025 cohort had in common

Age and scale. Klarna was founded in 2005, Chime and Figma in 2012, Circle in 2013; most listed with more than $500 million of revenue and after 10-plus years as private companies. The exception was CoreWeave, which pivoted from crypto mining to AI compute in 2019 and listed on the strength of Microsoft and OpenAI contracts. Venture-backed IPOs in 2025 were, in other words, catch-up listings of companies that would have gone public in 2021–2022 under earlier norms, plus one AI-infrastructure play.

What has 2026 delivered so far: SpaceX, xAI, Cursor and the AI IPO filings?

As of October 2026, the year holds the three largest venture-backed exits ever: SpaceX’s June IPO raised $75 billion at a valuation reported at about $1.7 trillion (its market value passed $2 trillion in early trading), the SpaceX–xAI all-stock merger valued xAI at $250 billion in Q1, and SpaceX agreed in June to acquire Cursor for $60 billion in stock.

The scale is hard to overstate. The PitchBook-NVCA Venture Monitor for Q2 2026 says the SpaceX listing alone created more exit value than all US venture exits of the prior decade combined, and that it was roughly 17 times the size of the next US venture-backed tech IPO of the year. Stout counts 186 US IPOs raising $134.2 billion in H1 2026, more than five times H1 2025, or $59.2 billion excluding SpaceX. Cerebras listed at about $34.3 billion, opened well above its offer price and later fell back below it, a reminder that AI-hardware listings trade on sentiment. In Q3, Bending Spoons listed in Europe at about $18.4 billion, Unitree Robotics at about $9 billion, Shein at $26.3 billion, and China’s CXMT raised $8.6 billion at an $85.5 billion valuation (state-backed, so not a classic venture exit).

M&A followed the same AI logic. The xAI merger (announced in early February 2026, with SpaceX valued at $1 trillion and the combined entity at $1.25 trillion) and the Cursor deal (an option agreed in April with a $10 billion break fee, exercised days after the IPO) made SpaceX the largest acquirer of venture-backed companies in history. Nvidia’s $12.9 billion bid for Hugging Face, AMD’s announced $8.2 billion acquisition of World Labs and Stripe’s $7.5 billion purchase of OpenRouter were the other headline deals of Q3 2026. Anthropic and OpenAI both filed confidentially for IPOs in June 2026; neither had listed as of October, and each could become a trillion-dollar exit. The routes these deals took, direct acquisition, all-stock merger, licensing-plus-hiring and tender offers, are compared in our guide to IPO, M&A and secondary exits.

What did LPs actually get back: DPI by vintage?

Distributions have lagged exits badly: Carta’s Q1 2026 analysis shows median realised DPI just above zero for 2019 and 2020 vintage funds, fewer than half of those funds having returned any capital, and only a small fraction of 2017–2018 funds reaching 1x DPI, even as paper TVPI rose for six straight quarters.

The mechanics explain the gap. Venture funds typically hold public stock through lock-ups and distribute or sell gradually; 2021’s $841.5 billion of exit value was followed by a 2022 crash that cut many of those positions by more than half before distribution. Then came three years in which the largest companies (Stripe, Databricks, SpaceX, OpenAI, Anthropic) chose tender offers and secondaries over listings, so their value stayed inside funds as unrealised marks. Carta’s Q4 2025 report covering 2,906 funds found that more than half of 2020-vintage funds had started generating some DPI, about a third of 2021-vintage funds, and just under a quarter of 2022 and 2023 vintages; median net IRR for 2021-vintage funds was about 1.4% and for 2022 about 0.7%, while top-decile funds across most vintages exceeded 20% IRR.

For LPs the practical lesson is that exit headlines and fund cash flows are different time series. The industry’s own figures (PitchBook puts 2025 global exit value at $549.2 billion) say liquidity has returned; the fund-level data say it has returned to a minority of funds. Our pillar on DPI, TVPI and the fund terms that govern distributions explains how to read those numbers in a manager’s reporting, and our guide to investing in venture funds as an LP covers access and minimums.

What do a decade of exits teach founders?

Three patterns hold across 2015–2026: the largest exits go to companies that stayed private long enough to reach scale, strategic acquirers pay the highest multiples for security and AI infrastructure, and the exit route is chosen by the market window rather than by the founder.

Scale first. Almost every top-of-year exit since 2019 involved a company more than eight years old with revenue above $500 million; SpaceX was 24 years old at listing. Founders who plan for an exit at Series C scale are planning for the long tail of M&A, which is where most exits by count actually happen, at prices rarely disclosed. Second, acquirers cluster: Google (Wiz, Mandiant, Looker), Microsoft (GitHub, Nuance, LinkedIn), Salesforce (Slack, Tableau), Cisco (AppDynamics, Splunk), Walmart (Jet, Flipkart) and now SpaceX, Nvidia and AMD have each paid record prices in categories adjacent to their core, which means founders in security, developer tools, AI infrastructure and commerce infrastructure have a deeper buyer pool than founders elsewhere. Third, timing: the gap between 2021 and 2022 shows that a six-month delay can halve an exit value, so processes should be run when the window is open.

Founders outside the US face an extra constraint. Europe’s largest exits remain trade sales rather than listings, as our analysis of why 92% of German exits are trade sales documents, and Japan’s venture market produces few billion-dollar outcomes at all; the regional firm landscape is covered in our review of the top venture firms in the UK and Europe. Browse the full Sectors & Exits by Year pillar for the sector reviews and the exit-route analysis.

Frequently Asked Questions