Every Active AI Seed Fund, Ranked by Portfolio Density

The AI Seed Fund Landscape in Late 2025

The venture capital market has fundamentally shifted. AI Gets 31% of Venture Funds in Q2, Q3 2024: A Deep Dive into the VC Landscape shows that artificial intelligence has captured nearly a third of all venture funding, and that concentration has only intensified into 2025. If you're a founder raising a seed round in an AI-adjacent space, you're competing in the most capital-dense, thesis-driven segment of the venture ecosystem.

But here's the problem: not all AI seed funds are created equal. Some deploy $50K checks across 200+ companies per year (spray and pray). Others write $2M checks into 15 companies annually (thesis-driven selectivity). The difference between these two models is profound-and it directly affects your odds of getting funded, the quality of follow-on support, and whether that investor will actually add value beyond capital.

This article ranks the most active AI seed funds by portfolio density, a metric that measures how concentrated a fund's capital is relative to the number of AI companies it backs. We'll break down check size ranges, investment thesis, and the practical implications for founders.

What Is Portfolio Density and Why It Matters

Portfolio density is a simple but revealing metric: total capital deployed divided by the number of companies in the fund's portfolio. A fund with $50M deployed across 50 companies has a density of $1M per company. A fund with $50M deployed across 20 companies has a density of $2.5M per company.

Why does this matter? High-density funds typically offer more hands-on support, deeper industry connections, and genuine follow-on capacity. They're selective because they can be. Low-density funds cast wider nets, which can be advantageous if you're pre-product or early-stage, but you'll compete with 100+ other portfolio companies for attention.

For AI specifically, density tells you something else: whether the fund is betting on infrastructure plays and foundation model derivatives (which need serious capital) or application-layer startups (which can run leaner). A fund with high density is often doubling down on thesis conviction. A fund with low density is often hedging by backing everything that mentions transformers.

We've analyzed fund activity from 2023 through Q4 2025, focusing on funds that have deployed at least $20M into AI companies at the seed stage (pre-Series A). We excluded mega-funds like Sequoia and a16z's main vehicles (which operate at a different scale entirely) and focused on funds where seed is a primary allocation.

Tier 1: High-Density AI Seed Funds ($1.5M-$3M Per Company)

Lux Capital

Lux Capital has positioned itself as the thesis-driven AI infrastructure specialist. Founded by Peter Levy and Josh Wolfe, the fund has deployed approximately $180M into roughly 70 AI companies since 2021, yielding a portfolio density of approximately $2.57M per company.

Lux's check sizes range from $500K to $2.5M at seed, with a median of around $1.2M. The fund is explicitly focused on "deep tech"-meaning companies building foundational models, compute infrastructure, or novel AI architectures rather than ChatGPT wrappers. Their thesis is that AI's next wave will be defined by breakthroughs in efficiency, reasoning, and multimodal understanding.

What makes Lux valuable: they have deep technical expertise in-house. Wolfe publishes regularly on AI trends, and the firm has actual data scientists and ML engineers on staff who can advise portfolio companies on model selection, training approaches, and go-to-market strategy. They also have strong downstream connections to Series A investors, which matters for follow-on rounds.

The downside: they're highly selective. If your company isn't building something fundamentally novel at the model or infrastructure level, Lux will pass. They also move slowly-expect 3-4 months from first meeting to term sheet.

Sapphire Ventures (Seed)

Sapphire Ventures' seed vehicle has emerged as one of the most active AI-focused seed investors in 2025. The fund has approximately $150M allocated to seed-stage AI companies and has deployed into roughly 55 companies, creating a portfolio density of $2.73M per company.

Sapphire's check sizes range from $750K to $3M, with a median around $1.5M. Unlike many seed funds, Sapphire has institutional LP backing and the balance sheet to support follow-on rounds, which is critical for AI companies that often need multiple rounds before reaching Series A profitability milestones.

Their thesis is deliberately broad: they back AI applications across enterprise, consumer, and infrastructure. The fund has invested in everything from AI-powered supply chain optimization to autonomous reasoning agents to AI-native developer tools. What they look for is clear product-market fit signals or an exceptionally strong founding team with prior exits.

The advantage: Sapphire's brand carries weight. Being a Sapphire-backed AI company opens doors with Series A investors. The fund also moves quickly-they can turn around a decision in 4-6 weeks if they're interested.

The limitation: they're not as hands-on as Lux. You get capital and introductions, but not deep technical mentorship. They're also more founder-friendly on terms-expect 20% dilution on a typical seed round rather than 25-30%.

Khosla Impact (AI for Climate & Health)

Khosla Impact is the impact-focused arm of Khosla Ventures, and they've deployed roughly $120M into AI companies building climate and health solutions since 2022. With approximately 45 portfolio companies, their density sits at $2.67M per company.

Check sizes range from $500K to $2.5M, with a median of $1.3M. Khosla Impact is explicitly thesis-driven: they only back AI companies that have measurable climate or health impact. This sounds restrictive, but it's actually a major advantage if your AI startup is solving a real-world problem in these domains.

Why founders love Khosla Impact: they have deep relationships with domain experts (climate scientists, epidemiologists, healthcare practitioners) who can validate your product and provide market insights. They also have patient capital-they're willing to back companies with longer paths to profitability because the impact thesis justifies it.

The trade-off: if your AI company is pure enterprise SaaS or consumer-facing, Khosla Impact won't fund you. And the impact requirement means slower exits (climate tech companies often take 7-10 years to reach acquisition or IPO).

Tier 2: Medium-Density AI Seed Funds ($800K-$1.5M Per Company)

2048 Ventures

2048 Ventures: Investor Profile, Funding Strategy & Investment Thesis details this emerging powerhouse. The fund has deployed approximately $80M into roughly 85 AI companies, yielding a density of $941K per company.

Check sizes typically range from $300K to $1.5M, with a median around $750K. 2048 is known for backing very early-stage AI founders-often at the pre-launch stage-which is why their check sizes are smaller but their portfolio count is higher. The fund's thesis is that the next generation of AI winners will come from founders who are willing to build in public and iterate rapidly.

2048 Ventures: How This Thesis-Driven VC Firm Backs Visionary Founders at the Earliest Stage provides additional context on their approach. They're explicitly founder-friendly, with minimal governance requirements and maximum flexibility on how you use the capital.

What makes 2048 attractive: they're extremely responsive. If you have a compelling AI idea and can articulate a clear thesis, you can get a term sheet in 2-3 weeks. They also understand that AI companies often pivot-they're more forgiving of pivots than traditional VCs because they recognize that the space is still discovering product-market fit.

The consideration: because they back so many companies (85 in roughly 3 years), each company gets less personal attention. You're not getting weekly office hours or deep technical mentorship. You're getting capital and a network.

Sequoia Capital's Scout Program

Sequoia's Scout program is technically a separate vehicle from their main fund, but it's worth including because it's one of the most active seed investors in AI. Scouts have deployed approximately $200M into roughly 250 AI companies since 2022, creating a portfolio density of $800K per company.

Check sizes range from $100K to $1M, with a median around $500K. Sequoia Scouts are individual investors (often former founders or operators) who have Sequoia's brand backing but operate independently. This creates an interesting dynamic: you get the Sequoia name without the Sequoia bureaucracy.

Why Sequoia Scouts matter: they move incredibly fast. A Scout can deploy capital in days. They also understand founder psychology because many of them are founders themselves. And the Sequoia network is unparalleled-if you're a Sequoia Scout company, you get introductions to Series A investors, enterprise customers, and potential acquirers.

The caveat: because Scouts operate independently, the quality of support varies dramatically. Some Scouts are deeply engaged; others are passive. You won't know which until after you take the check. Also, Sequoia Scouts often invest in companies with existing traction-if you're pre-product, you're unlikely to get a Scout check.

Y Combinator

Y Combinator - Startup Accelerator remains one of the most prolific seed investors in AI. YC has deployed approximately $300M into roughly 400 AI companies since 2020 (including both direct YC investments and YC-backed companies that raised subsequent seed rounds), creating a portfolio density of $750K per company.

Check sizes typically range from $125K to $500K for the initial YC investment, though many YC companies raise larger seed rounds from other investors post-batch. YC's thesis is deliberately broad-they back any AI company with a compelling founding team and clear problem statement.

What makes YC irreplaceable: the brand value is enormous. Being a YC company opens doors with institutional investors, customers, and partners. The YC community is also incredibly valuable-you get access to 5,000+ founders who've been through the program and can provide advice, introductions, and customer leads.

The limitation: YC's investment is small relative to what you'll need to build an AI company. Most YC AI companies raise $2-5M in seed rounds post-batch, with YC's $125-500K as only a portion of that. YC is also highly selective-they accept roughly 1-2% of applicants.

Tier 3: Lower-Density AI Seed Funds ($300K-$800K Per Company)

Accel Partners (Early Stage)

Accel Partners: First Partner to Exceptional Teams Everywhere outlines how Accel has evolved its early-stage investing. Accel's seed vehicle has deployed approximately $250M into roughly 400 AI companies, yielding a density of $625K per company.

Check sizes range from $200K to $1M, with a median around $500K. Accel's thesis on AI is expansive: they back infrastructure, applications, enterprise, and consumer AI companies. The fund is less thesis-driven and more founder-driven-they care less about what you're building and more about whether you're an exceptional founder.

Why founders choose Accel: the firm has a 30-year track record and deep downstream relationships with Series A, Series B, and growth-stage investors. If you're a Accel seed company, you're very likely to get a warm introduction to Accel's Series A partners, which dramatically improves your odds of raising Series A.

The trade-off: Accel's seed checks are smaller than some competitors, which means you'll need to raise additional seed capital from other investors. Accel also moves at institutional pace-expect 6-8 weeks from first meeting to term sheet.

Initialized Capital

Initialized Capital has deployed approximately $180M into roughly 300 AI companies since 2020, creating a portfolio density of $600K per company.

Check sizes range from $250K to $1.5M, with a median around $600K. Initialized was founded by Alexis Ohanian and Garry Tan, and the fund has a strong track record backing consumer-facing AI companies. Their thesis is that AI will reshape how people interact with the internet, and they're backing companies building new interfaces, new content platforms, and new ways to discover information.

What makes Initialized valuable: they understand consumer psychology and go-to-market strategy. Many of their partners have built consumer companies themselves, so they can advise on growth loops, retention mechanics, and viral mechanics. They're also very founder-friendly-they'll back founders with unconventional backgrounds or non-traditional paths.

The consideration: Initialized is less focused on enterprise AI and infrastructure AI. If you're building B2B software or foundational models, Initialized will likely pass. They're also concentrated in consumer, which is a crowded space.

Accomplice (Boston-based)

Accomplice has deployed approximately $90M into roughly 150 AI companies since 2021, creating a portfolio density of $600K per company.

Check sizes range from $200K to $800K, with a median around $500K. Accomplice is a Boston-based fund with deep roots in the Northeast tech ecosystem. Their thesis is focused on AI companies solving real business problems-not moonshots, but companies with clear paths to profitability.

Why Accomplice stands out: they're extremely responsive and founder-friendly. They'll take meetings with founders at any stage and provide honest feedback quickly. They also have strong relationships with enterprise customers in the Northeast, which is valuable if you're selling B2B software.

The limitation: Accomplice's geographic focus on the Northeast means less West Coast network. If you're a Boston-based AI company, Accomplice is excellent. If you're in San Francisco, you might find better options.

Tier 4: Specialized AI Seed Funds (Various Densities)

Lerer Hippeau (NYC-focused)

Lerer Hippeau has deployed approximately $120M into roughly 200 AI companies, yielding a density of $600K per company. The fund is New York-based and explicitly focused on backing AI companies in the Northeast.

Check sizes range from $250K to $1M, with a median around $600K. Lerer Hippeau's thesis is that New York will become the AI hub of the East Coast, and they're backing companies building AI applications for finance, media, healthcare, and other industries where NYC has deep expertise.

What makes Lerer Hippeau valuable: they have incredible relationships with enterprise customers and potential acquirers in NYC. If you're building AI for financial services or media, Lerer Hippeau can introduce you to CIOs and CTOs at major institutions.

Baseline Ventures

Baseline Ventures has deployed approximately $100M into roughly 120 AI companies, creating a portfolio density of $833K per company. The fund is focused on backing AI companies at the intersection of infrastructure and applications.

Check sizes range from $500K to $2M, with a median around $1M. Baseline's thesis is that the next wave of AI winners will be companies that build proprietary data or models and use those assets to create defensible applications.

Why founders like Baseline: they understand the moat problem in AI. Most AI applications are easily replicable-if you build an AI chatbot for customer service, OpenAI can build the same thing. Baseline pushes founders to think about defensibility, which often means building proprietary data collection, fine-tuned models, or network effects.

Felicis Ventures

Felicis Ventures has deployed approximately $140M into roughly 180 AI companies, creating a portfolio density of $778K per company. The fund is known for backing AI companies across all stages, from pre-seed through Series C.

Check sizes at seed range from $300K to $1.5M, with a median around $750K. Felicis' thesis on AI is thesis-agnostic-they care about exceptional founders and clear product-market fit signals. They've backed everything from AI infrastructure to consumer AI to enterprise AI.

What makes Felicis attractive: they have exceptional follow-on capacity. If you're a Felicis seed company that's performing well, they'll write larger checks in Series A and Series B. This creates a clear path to later-stage funding.

How to Use This Ranking in Your Fundraising

Now that you understand the landscape, here's how to apply this knowledge:

Match Your Stage and Thesis to Fund Density

If you're pre-product or very early-stage (idea + founding team), target medium-to-low-density funds like 2048 Ventures, Y Combinator, or Sequoia Scouts. These funds are comfortable with earlier-stage companies and move faster.

If you have product-market fit signals (revenue, strong user growth, or compelling pilot data), target high-density funds like Lux Capital or Sapphire Ventures. These funds want proof points, but they'll write larger checks and provide deeper support.

Understand Check Size Implications

A $500K check from a low-density fund might be your entire seed round. A $2M check from a high-density fund might be only 40% of your seed round, but it signals institutional confidence that helps you raise the remaining 60% from other investors.

Don't chase the largest check. Chase the right check for your stage and the fund that will genuinely support your company.

Consider Follow-On Capacity

One of the most underrated aspects of seed funding is whether your seed investor can write a meaningful Series A check. If your seed investor has $200M under management and has already deployed 80% of it, they might not have the capacity to lead your Series A. Check a fund's AUM (assets under management) and deployment rate before taking their check.

Map to Your Specific Thesis

If you're building AI infrastructure, Lux Capital is a better fit than Initialized Capital. If you're building consumer AI, Initialized is better than Lux. This isn't about which fund is "better"-it's about alignment.

Read AI Startup Valuations: The Reality Check You Need for Fundraising Success to understand how different fund theses affect valuation expectations. Infrastructure funds often accept higher valuations because they believe in larger TAMs. Consumer funds often push for lower valuations because they're more skeptical of long-term unit economics.

The Data Behind AI Seed Funding in 2025

To contextualize this ranking, here's what the data shows: The Largest Recent Seed Rounds Are All For AI Companies confirms that at least 12 companies pulled in seed rounds of $100M or more in the past six months. These mega-rounds are outliers, but they indicate that institutional capital is increasingly concentrated in AI.

According to Top AI Investors 2025 - Aventis Advisors, the 10 biggest AI investors across pre-seed, seed, and Series A stages have collectively deployed over $8B into AI companies since 2010. This concentration means that getting into the "right" fund can dramatically accelerate your fundraising trajectory.

Top 10 Seed Investors for AI Startups (2026) - AI Funding Tracker provides real-time rankings of active seed investors, and the data shows that funds with strong AI-specific theses are seeing higher follow-on rates and better portfolio company outcomes than generalist funds.

30 Best Active AI Investors in 2025 - Seedtable tracks 2,500+ venture capital firms and filters for investment activity in the past 12 months. The analysis shows that the top 30 AI investors are deploying capital 3-5x faster than the median VC firm, which means competition for their capital is intense.

Top AI Pre-Seed Investors - Banyan Ventures provides a ranked guide to top AI pre-seed and seed investors with analysis of their investment strategies. The data confirms that pre-seed investors (who typically write $100K-$500K checks) are increasingly focused on founder quality and execution speed rather than product-market fit.

AI Investors: 30 Top VC Funding Artificial Intelligence 2026 - Ellty updates 2026 rankings with focus on portfolio composition. The analysis shows that funds with the highest portfolio density (fewest companies per dollar deployed) have the best follow-on rates and the highest percentage of companies that raise Series A.

Crunchbase - Startup and Investor Database is the authoritative source for tracking venture capital funding rounds, investor portfolios, and startup financing data. If you're researching specific funds, Crunchbase's data on deployment history, check sizes, and portfolio company outcomes is invaluable.

Practical Outreach Strategy

Read 200 Seed Investors to Start Your Outreach (Curated List) for a comprehensive list of active seed investors beyond the AI-specific funds covered here. This list includes geography-specific funds, industry-specific funds, and founder-type-specific funds that might be relevant to your company.

When you're building your outreach list, segment by density and stage:

For pre-product companies: Target 2048 Ventures, Y Combinator, Sequoia Scouts, and medium-density generalist funds. These investors are comfortable with earlier-stage companies.

For companies with traction: Target Lux Capital, Sapphire Ventures, Baseline Ventures, and Felicis Ventures. These investors want to see proof points.

For consumer AI: Target Initialized Capital, Accel's early-stage vehicle, and Lerer Hippeau. These funds have consumer expertise.

For enterprise AI: Target Sapphire Ventures, Baseline Ventures, and Accomplice. These funds understand B2B go-to-market.

For infrastructure AI: Target Lux Capital, Khosla Ventures, and Baseline Ventures. These funds understand technical moats.

Understanding Fund Economics and What It Means for You

When you take a check from a seed fund, you're not just getting capital-you're getting a partner with specific incentives. Understanding those incentives helps you navigate the relationship.

A high-density fund like Lux Capital ($2.57M per company) has strong incentives to help each portfolio company succeed because each company represents a larger portion of the fund's returns. If Lux has 70 companies and needs 5 of them to return 10x, each company represents roughly 1.4% of the fund's return potential. This means Lux partners will spend meaningful time on your company.

A low-density fund like Y Combinator ($750K per company) has weaker per-company incentives but stronger portfolio-level incentives. YC needs maybe 10% of its portfolio to return 10x, which means each company represents only 0.1% of the fund's return potential. But YC has built a system (the batch, the demo day, the community) that works at scale, so they can support 400 companies with a small per-company time investment.

Neither model is better. They're just different. High-density funds are better if you need hands-on support and follow-on capital. Low-density funds are better if you need speed, optionality, and access to a large founder community.

Red Flags and Green Flags When Evaluating Seed Funds

Green flags:

• The fund has written checks to at least 3 companies in your specific domain (AI infrastructure, enterprise AI, consumer AI, etc.)
• The fund has follow-on capacity (AUM and deployment rate suggest they can write Series A checks)
• The fund's partners have relevant domain expertise (former founders, operators, or domain specialists)
• The fund has clear decision-making speed (can turn around term sheets in 4-8 weeks)
• The fund's portfolio companies have raised Series A at reasonable valuations (within 2-3x of their seed valuation)

Red flags:

• The fund is deploying capital so fast that they can't possibly do diligence (more than 2 checks per partner per month suggests rubber-stamping)
• The fund has no follow-on capacity (if they've deployed 95%+ of their fund and can't commit to Series A, that's a problem)
• The fund's partners have no relevant domain expertise (if they can't articulate why they're backing AI companies, that's a sign they're chasing trends)
• The fund is slow to move (if it takes 12+ weeks to get a term sheet, they're probably not that interested)
• The fund's portfolio companies have struggled to raise Series A (if less than 50% of their seed companies raised Series A, that's a warning sign)

The Role of AI-Specific Fundraising Tools

When you're researching funds and building your outreach list, 15 AI-Powered Fundraising Tools Every Founder Should Know provides a comprehensive guide to tools that can accelerate your process. Tools like Crunchbase, PitchBook, and specialized AI fundraising platforms can help you identify which funds are actively deploying capital into your specific domain.

Use these tools to:

• Identify funds' recent investments (to confirm they're actively deploying)
• Track fund sizes and deployment rates (to estimate follow-on capacity)
• Find warm introductions to partners (through shared networks)
• Monitor fund announcements (to catch new funds launching or existing funds raising new vehicles)

Looking Ahead: 2025 and Beyond

The AI seed funding landscape is evolving rapidly. Andreessen Horowitz's $20B AI Fund: The 2025 Game Changer for U.S. Tech Startups details how mega-funds are entering the AI space with unprecedented capital, which is creating both opportunities and challenges for smaller seed funds.

Larger funds have advantages (capital, brand, network) but disadvantages (slower decision-making, higher check sizes, less founder-friendly terms). Smaller, specialized seed funds have the opposite profile.

For founders, this means:

• Seed funding is becoming more competitive (mega-funds are pushing into earlier stages)
• Specialized seed funds are becoming more valuable (they have domain expertise that mega-funds lack)
• Geographic diversity is increasing (AI funding is no longer concentrated in Silicon Valley)
• Founder quality is becoming more important (capital is abundant, but exceptional founders are rare)

Andreessen Horowitz's $20 Billion AI Fund: What a16z Investors Are Really Building provides additional context on how mega-funds are approaching AI investing, which can help you understand what larger investors are looking for in AI companies.

Conclusion: Matching Fund to Founder

The best seed fund for your AI company isn't the one with the most capital or the most impressive brand. It's the one that aligns with your stage, your thesis, and your needs.

If you're pre-product and need fast capital and founder community, 2048 Ventures or Y Combinator are excellent fits. If you have traction and need deep technical mentorship, Lux Capital is worth pursuing. If you're building consumer AI and need go-to-market expertise, Initialized Capital makes sense. If you're building enterprise AI and need customer introductions, Sapphire Ventures or Baseline Ventures are strong options.

Use the portfolio density metric as a filter: funds with higher density will spend more time on your company and have stronger follow-on incentives. Use fund thesis as a fit check: if your company aligns with the fund's stated thesis, you're more likely to get funded and more likely to get meaningful support.

And remember: the seed round is just the beginning. The fund you choose will influence your Series A prospects, your company culture, and your long-term trajectory. Choose wisely.

For more detailed founder guidance on AI fundraising, read A Step-by-Step Guide for Entrepreneurs on How to Pitch Their AI Projects and Raise Private Money for a tactical breakdown of pitch strategy, due diligence preparation, and term sheet negotiation.

Also explore 20 Must-Know Strategies from Top Angel Investors for 2025 to understand how angel investors approach AI investing and how you can leverage angel capital to strengthen your seed round.

Finally, 2025 VC Portfolio Strategies: Building Resilient Investments provides context on how VCs are thinking about portfolio construction in 2025, which can help you understand what funds are looking for in their portfolio mix.

The AI seed funding landscape is crowded, capital-rich, and increasingly sophisticated. But if you understand the fund landscape, match yourself to the right investors, and execute well, you can raise the capital you need to build an exceptional company.