This article is republished from The Paper · IT Juzi, all rights reserved by the original authors. Click to view the original link.

---

In the First Half of the Year, VC Money Was Half Invested in AI; These 30 Companies Alone Raised Over 170 Billion Yuan

Author | Wu Meimei Source | IT Juzi

In the first half of 2026, Chinese AI startups experienced a frenzied fundraising boom, with total funding exceeding 300 billion yuan. The volume raised in just 6 months surpassed the entire year of 2025.

AI funding accounted for approximately 48.6% of the market, nearly half, while the number of deals accounted for about 22.5%.

In other words—in the first half of 2026, for every 2 yuan raised in the primary market, nearly 1 yuan flowed to AI. Although the number of deals was only over 20%, the amount accounted for nearly half, indicating that the average deal size in the AI sector is far higher than in other sectors.

More significant than the total funding volume is the structural shift in capital flows. Our core findings include:

① The large model sector is the most lucrative, with funds highly concentrated in top players. DeepSeek, StepFun, and Moonshot AI (Kimi) collectively raised approximately 93 billion yuan, accounting for 30% of the total market funding. Capital is heavily betting on top players with a "winner takes all" logic.

② World Models have become the biggest consensus in early-stage investment. Six early-stage startups focusing on world models collectively raised approximately 9.7 billion yuan, with capital preemptively positioning itself in the "operating system layer" of embodied intelligence.

③ The AIGC application sector is the most mature in terms of commercial implementation of AI technology, represented by image & video generation.

I. AI Sector Sprinting in the First Half, Funding Surges

From January to June 2026, data from IT Juzi shows that equity financing in the domestic AI sector exhibited core characteristics of "high total growth and explosive increase in funding amounts." Capital investment in AI continued to intensify, with the scale of financing in just 6 months far exceeding the level of the entire year of 2025.

According to IT Juzi data, there were 1,203 AI sector financing events in the first half of 2026, with total funding breaking through 300 billion yuan, far exceeding the entire year of 2025. This reflects sustained strengthening of investor confidence in the AI sector, indicating that the AI industry has fully transitioned from the technology verification stage to an explosive period of scaled implementation.

The financing rhythm in the AI sector in the first half of 2026 maintained high levels starting from March; domestic AI financing scales remained above 40 billion yuan in March, April, and May; June saw explosive growth, with monthly total funding breaking through 100 billion yuan, mainly due to the completion of DeepSeek's 51 billion yuan first-round financing, continuously pushing up the overall financing scale of the sector.

The regional landscape of AI sector financing in the first half of 2026 showed a pattern of "Beijing, Hangzhou, Shanghai, and Shenzhen absolutely leading, with significant agglomeration effects in the Yangtze River Delta region, and new first-tier cities accelerating follow-up," highlighting prominent characteristics of industrial cluster development.

In the AI industry financing landscape of the first half, Beijing, Shanghai, Hangzhou, and Shenzhen absolutely led, contributing 73.89% of the total number of financing transactions and monopolizing 86% of the financing scale, becoming the absolute core of China's AI financing.

Among them, Beijing ranked first nationwide with 321 financing events and a total financing amount of 95.517 billion yuan. Hangzhou, benefiting from DeepSeek's explosion and the first round of over 50 billion yuan subscription, surged past Shanghai and Shenzhen to rank second nationwide. Shanghai had 211 financing events and a total financing amount of 59.570 billion yuan in the AI industry in the first half, while Shenzhen had 215 financing events and a total financing amount of 35.943 billion yuan. Suzhou ranked fifth nationwide with 19.022 billion yuan, becoming the third pole of the Yangtze River Delta.

II. Financing Scale of Various AI Sub-sectors

Looking at the distribution characteristics of sub-sectors: in terms of total funding distribution, the large model sector dominates alone, taking more than half of the funds across all sectors.

The large model sector became the absolute core technology foundation of the AI track with a total financing amount of 159.853 billion yuan. Single-project hundred-billion-level financings frequently appeared in top projects, showing the strongest Matthew effect of funds, making it the first heavy-position sector in the primary market.

Additionally, in the first half of 2026, the total financing for the AI infrastructure layer was 72.568 billion yuan. Computing power, AI chips, and training frameworks belong to asset-heavy tracks with frequent large financings; the AI technology layer leans towards simulation and spatial algorithms, with fewer financing events and lower funds compared to the infrastructure layer. These two major sectors combined had 258 events and a total financing of 106.783 billion yuan, accounting for 18.00% of the aggregated total funds.

The AI + Embodied Intelligence sector is the second growth pole, with a total financing amount reaching 90.644 billion yuan and 312 financing events, making it the sector with the most financing events and highest project activity across all tracks.

The AIGC application sector had a total financing of 59.605 billion yuan in half a year, being the most mature sector for commercial implementation of AI technology, with the highest recognition of its commercial value by capital.

From the perspective of single-project financing capability, large models and AIGC are the most outstanding. The three sectors with the highest average financing amount per deal are: AIGC (710 million yuan) and Large Models (704 million yuan), both far exceeding the average across all sectors.

III. Core Betting Rhythm of Capital

The capital layout in the AI sector in the first half of 2026 presented a clear rhythm of "heavy positions in growth stage, ballast in maturity stage, incubation in early stage," with significant differences in financing scale, number of events, and capital preferences across the three investment stages.

1. Early Stage: Innovation Source and Incubation Pool of the Sector

According to IT Juzi statistics, the TOP 10 early-stage companies secured a total of 16.589 billion yuan, accounting for 36.7% of total ultra-early-stage financing.

• Insight 1: World Models become the "First Consensus" of early-stage investment. Six companies in the world model direction collectively raised 9.7 billion yuan (accounting for 58%). World models are viewed as the "operating system" of embodied intelligence—whoever masters the world model masters the robot's ability to understand the physical world.
• Insight 2: Significant "inflation" in angel round financing amounts. Among the TOP 10, four companies completed financings exceeding 700 million yuan in seed or angel rounds, which was unimaginable before 2024.
• Insight 3: Spillover effect of big-company talent ecosystems emerges. Big company AI talents and technologies are spreading outward through incubation/entrepreneurship, forming a "big-company lineage" startup ecosystem.
• Insight 4: No pure large model companies in the early-stage TOP 10. The early window for the large model sector has basically closed, with capital shifting to "downstream of large models"—world models, embodied brains, and physical AGI.

2. Growth/Maturity Stage: Core Targets Heavily Bet On by Capital

According to IT Juzi statistics, the TOP 20 companies raised a total of 156.5 billion yuan, accounting for more than half of the total financing across all sectors.

• Camp 1: The Three Giants of Large Models (Total 93.006 billion yuan, accounting for 59.4% of TOP 20): DeepSeek, StepFun, and Kimi swept the top three spots, each securing over 10 billion yuan in financing; notably, DeepSeek obtained 51 billion yuan in just one Series A round, making it the largest financing event in the AI sector in the first half of 2026.
• Camp 2: The "Seven Samurai" of Humanoid Robots (Total 28.226 billion yuan, accounting for 18.0% of TOP 20): Zibianliang Robot, Zhifangping, Qianxun Intelligent, Jijia Vision, Xinghaitu, Galaxy General Robot, and Xingdong Era all secured large financings exceeding 2 billion yuan, with financing stages collectively entering Series B to strategic investment.
• Camp 3: The "Three Musketeers" of AIGC Applications (Total 6.778 billion yuan, accounting for 4.3% of TOP 20): Shengshu Technology, Sand.ai, LiblibAI, and Aishi Technology focus on multimodal video/image generation and have preliminarily achieved scaled commercial implementation.
• Camp 4: Autonomous Driving & Others (Total 25.893 billion yuan, accounting for 16.6% of TOP 20): Momenta and Kuwa Technology are in the Pre-IPO stage; Huashen Zhiyao became the largest single financing in the AI drug discovery sector with 5.533 billion yuan; XiWang Sunrise's 4 billion yuan strategic investment represents the logic of domestic substitution for AI chips.

IV. Outlook for the Second Half

Standing at the midpoint of 2026, we predict that the annual financing scale will likely break through 600 billion yuan, but the rhythm in the second half will be "high in the front, low in the back."

The total financing of 307.682 billion yuan in the first half has already established a high base. According to seasonal patterns, financing is usually more active in the second half, so breaking through 600 billion yuan for the whole year is a probable outcome.

Furthermore, the large model sector will welcome its first "elimination tournament." The three giants took away 93 billion yuan in the first half, and Zhipu and MiniMax have entered the secondary market, leaving a rapidly narrowing funding window for the remaining 200+ model companies. We expect the first cases of layoffs/transformation/M&A among large model startups to appear in the second half. Surviving companies must either find clear differentiated scenarios or secure strategic investments from industrial giants. The entrepreneurial window for purely "general-purpose large models" has closed.

—END—

Original Title: "In the First Half of the Year, VC Money Was Half Invested in AI; These 30 Companies Alone Raised Over 170 Billion Yuan"