Back to News
IndustryAI Understanding briefing

DeepSeek closes $16 billion funding round at $74 billion valuation

DeepSeek has closed a second funding round of approximately $16 billion, more than doubling its initial target and reaching a pre-money valuation of $74 billion, with significant participation from state-owned institutions and domestic hardware manufacturers.

4 min readRead the linked source
Source-provided image accompanying DeepSeek closes $16 billion funding round at $74 billion valuation
Source referenceSource recorded
Publisher
oodaloop.com
Source type
Linked source — primary-source status has not been established.
ContextUnderstand this in 60 seconds

Key terms

Memory (Agent Memory)
Stored context an AI agent uses across steps or sessions to improve continuity.
Compute
The processing resources required to train and run models, often measured in FLOPS or GPU hours.
Prompt
The input instructions and context provided to a generative model.

What happened

According to OODA Loop, Chinese AI startup DeepSeek closed a second funding round of roughly 100 billion yuan (approximately $16 billion), which is more than double its initial target. This round brings DeepSeek's pre-money valuation to $74 billion. The financing includes substantial participation from state-owned institutions and government-backed funds, which accounted for 27% of the round, marking a significant increase in state involvement compared to previous financing. Existing private backers such as Tencent and CATL remained leading investors, while new participants include domestic hardware and memory players like SMIC and ChangXin Memory. The report states that DeepSeek doubled the size of the raise to build a financial buffer against high black-market prices for advanced Nvidia hardware, which is subject to U.S. export controls. To maintain operational agility, the company structured major equity stakes through founder-controlled vehicles with strict long-term lock-up provisions.

DeepSeek closed a second funding round of approximately 100 billion yuan, or nearly $16 billion, according to OODA Loop. This amount is more than double the company's initial financing target. The round establishes a pre-money valuation of $74 billion for the Chinese AI startup.

State-owned institutions and government-backed funds contributed 27% of the total round, a significant increase from previous financing rounds. This shift indicates a deeper integration of state capital into DeepSeek's ownership structure, aligning the company more closely with national strategic goals.

Prominent private investors, including Tencent and CATL, continued to lead the round. New participants included SMIC and ChangXin Memory, key domestic players in semiconductor manufacturing and memory production. Their involvement is reported to be a strategic response to U.S. export controls on advanced hardware.

The company reportedly doubled the size of the capital raise to create a financial buffer against the high costs of acquiring advanced Nvidia hardware on the black market. To preserve operational flexibility, DeepSeek structured significant equity stakes through founder-controlled vehicles that include strict long-term lock-up provisions.

Source details: oodaloop.com

Why it matters

This funding round represents a major shift in the financial structure of China's leading AI companies, highlighting the increasing role of state capital in supporting critical national AI champions. The explicit inclusion of domestic hardware manufacturers like SMIC and ChangXin Memory signals a strategic move to mitigate the impact of U.S. export controls on advanced computing hardware. By securing a massive financial buffer, DeepSeek aims to navigate the scarcity and high cost of high-performance Nvidia chips, which are essential for training and deploying large-scale AI models. This development underscores the intensifying geopolitical competition in AI, where access to resources is a primary bottleneck. The heavy state backing suggests that DeepSeek's development is now closely aligned with national strategic interests, potentially influencing its research priorities and deployment strategies. For the global AI industry, this move indicates that Chinese AI firms are becoming more financially resilient and less dependent on external hardware markets, which could alter the competitive landscape in the coming years.

The substantial increase in state-backed funding highlights the Chinese government's commitment to supporting its domestic AI champions. This financial backing provides DeepSeek with the resources to compete globally despite hardware restrictions.

The participation of domestic hardware manufacturers like SMIC and ChangXin Memory is a critical development. It suggests a coordinated effort within China's tech ecosystem to develop a self-sufficient supply chain for AI , reducing vulnerability to external sanctions.

The high valuation of $74 billion reflects investor confidence in DeepSeek's technology and market position. It also signals that the AI sector in China is attracting significant capital, which could accelerate innovation and deployment of AI models.

The use of founder-controlled vehicles with lock-up provisions allows DeepSeek to maintain strategic autonomy while accepting large-scale state and corporate investment. This structure may influence how the company navigates future regulatory and geopolitical challenges.

Interactive Mechanism

Interactive Mechanism: How It Actually Works

Explore the underlying technology behind this development interactively.

Document Size:128K tokens
Needle Placement Depth (Location in document):50% into text
Attention Context Buffer Map:
Target Fact (50%)
Equivalent Pages~320Standard book pages
Retrieval Accuracy99.9%Needle recall score
RAM / KV Cache5.1 GBMemory overhead
Prompt CachingActive~80% discount on reuse
Core takeaway: Million-token context windows allow querying whole codebases or legal archives in one prompt. However, KV cache memory scales with context length, making prompt caching crucial for real-time production.
Interactive Concept Check+10 Points
AI Models Explained Quiz

Which component of an AI application is the machine-learning model itself?

What to watch next

Investors and analysts should monitor how DeepSeek utilizes this capital to secure hardware and expand its model capabilities. The extent to which state-backed funds influence DeepSeek's strategic decisions and research directions will be a key factor in understanding the future trajectory of Chinese AI development. Additionally, the reaction of U.S. policymakers and the broader tech industry to this increased state involvement in DeepSeek's capital structure may lead to further regulatory or export control measures. The performance of domestic hardware suppliers like SMIC and ChangXin Memory in supporting DeepSeek's needs will also be critical to assess the effectiveness of China's efforts to reduce reliance on foreign technology.

The allocation of the $16 billion in funding will be a key indicator of DeepSeek's strategic priorities. Whether the capital is directed toward hardware acquisition, model development, or global expansion will shape its competitive position.

The impact of U.S. export controls on DeepSeek's ability to access advanced hardware remains a critical variable. The success of domestic alternatives in meeting DeepSeek's needs will determine the long-term sustainability of its growth.

The role of state-backed funds in DeepSeek's governance and decision-making processes may evolve. Increased state influence could lead to a focus on national security and strategic applications, potentially diverging from purely commercial objectives.

The reaction of the global AI community to DeepSeek's funding round may influence international perceptions of Chinese AI capabilities. It could also further discussions on export controls and technology competition between the U.S. and China.

Related guides & quizzes

Found this useful?