Khosla Ventures is in talks to raise up to $5.5 billion across a new family of funds, which would mark the venture firm's largest capital haul in its 20-year history since 2004, according to Bloomberg (reported July 23, 2026). The firm is considering allocating roughly $1 billion for seed-stage startups, $2 billion for early-stage venture rounds, and $2.5 billion for its opportunity fund targeting more mature companies. The proposed split represents a deliberate bet on seed and early-stage, an unusual posture at a time when most megafunds are concentrating on late-stage foundation models.
Khosla Ventures' July dealmaking reinforces this thesis. In July 2026 alone, the firm led a $120 million Series C for legal AI startup Norm AI (valued at $1.2 billion) and a $300 million Series A for quantum computing startup Oratomic. In June, it backed General Intuition's $320 million Series A (training AI on gameplay data) just three months after leading its seed. The firm's track record as OpenAI's first outside investor—a $50 million check in 2019 at $1B valuation, now valued at ~$1.5B—provides credibility for this bet on finding the next generational company before consensus prices it in.
The $5.5B target, if achieved, would represent a ~40% increase from Khosla's previous fundraise close of approximately $4 billion in 2025. This follows a broader pattern: Founders Fund closed $6 billion in 2026 for concentrated late-stage AI and deep-tech bets, Sequoia raised ~$7 billion for an expansion fund, and Andreessen Horowitz deployed >$15 billion across six funds in early 2026. Bloomberg and sources cautioned that the final figure could shift, but the direction signals continued investor confidence in AI venture despite market volatility.
For founders, Khosla's fund size matters because it translates to larger check sizes in follow-on rounds—important when a company needs $200M+ Series A or B without investor hesitation. For the ecosystem, the emphasis on seed and early-stage is notable: rather than competing for consensus winners in the frontier model arms race, Khosla is betting that AI's value will disperse into applied software, regulated industries, robotics, and the physical world, and that the next tier of category-defining companies remains unfunded at seed. The $3B allocated to seed and venture contrasts sharply with the late-stage-heavy posture of competitors.