DeepSeek funding round nears $12B as Tencent and CATL commit

DeepSeek funding has escalated from ambitious to historic. The Chinese AI startup is close to securing at least 80 billion yuan, roughly $11.93 billion, in a new financing round, according to Bloomberg News, which cited people familiar with the matter. The raise would rank among the largest private financings ever assembled by an AI company, and it arrived after investor demand blew past the company's original target.

The DeepSeek funding round has drawn some of China's most consequential companies into the cap table. Tencent Holdings and the battery maker CATL have committed among the largest amounts in the current round, according to Bloomberg News. The pairing of an internet giant with an energy and battery giant signals how the financing blends AI ambition with the power and hardware interests that underpin large-scale model training.

Investor appetite overshot the plan. DeepSeek initially sought about 50 billion yuan, but demand exceeded expectations following the successful launch of its latest AI model, according to Bloomberg News. The final tally could approach 100 billion yuan, a figure that would push the DeepSeek funding effort toward roughly $15 billion at current exchange rates.

None of the three named companies responded to requests for comment. Reuters, which reported Bloomberg's account on Tuesday, said it could not immediately verify the report.

What the DeepSeek funding round reveals about AI demand

Financings at this scale do not happen for frontier research alone. They happen when investors believe the model pipeline is already strong enough to justify an enormous compute buildout. The DeepSeek funding story fits that pattern: a raise this large is effectively a bet on the company's ability to keep shipping models that compete globally while training on a hardware stack that China controls.

That context matters because DeepSeek has recently leaned into domestic hardware. The company partnered with Huawei Technologies to develop programming tools optimized for Huawei's Ascend AI chips, according to Reuters, highlighting China's push to reduce reliance on Nvidia's AI ecosystem. A funding round this large gives DeepSeek the capital to keep investing in that alternative stack.

The broader AI funding landscape makes the raise look less like an outlier and more like a milestone. Enterprise AI labs have been raising ever-larger rounds to fund compute and talent, from Cohere's enterprise agent push to mid-stage AI agent financings, a trend covered in recent reporting on AI agent infrastructure funding. The DeepSeek funding round now sets the high-water mark for the current wave.

Model momentum behind the raise

The financing follows the recent release of DeepSeek-V4.1-Flash. The company said the model was designed for greater capability, faster inference, higher throughput and scaling to larger models, according to Reuters. Strong reception for that release reportedly helped lift the round beyond its original target, a reminder that in AI, model launches double as fundraising events.

DeepSeek's ascent has been unusually compressed. The company began accepting outside capital only recently by industry standards, yet it is now closing one of the largest private AI rounds ever recorded. For the AI agent ecosystem, where startups like Cohere have been building enterprise agent platforms, the DeepSeek funding news resets expectations about how quickly a lab can move from research project to institutionally backed contender.

Observers of the financing note the unusual mix of strategic investors. Internet platforms bring distribution and cloud infrastructure. A battery manufacturer brings energy. In a world where training runs are constrained as much by power as by chips, that combination is more strategic than it looks at first glance.

Chips, energy and the IPO runway

Beyond the round itself, DeepSeek appears to be preparing for life as a public company. The company hired CITIC Securities to prepare for a potential STAR Market initial public offering, according to Reuters, which reported the move last month. The timing, valuation and fundraising size for the listing remain undecided.

The DeepSeek funding round lands at a moment of intense capital formation in AI. Western labs have raised tens of billions from cloud providers and sovereign funds; Chinese labs have historically raised less, constrained by different capital markets. A $12 billion-plus round closes that gap decisively and signals that domestic Chinese capital is willing to fund frontier AI at American scale.

There are still open questions. The round has not closed, the figures remain sourced to anonymous people familiar with the matter, and the company has not commented. But if the financing completes at anything near the reported size, it will reshape the map of AI funding — and confirm that the race between open-weight and frontier labs is now being fought with war chests, not just weights.