Portuguese 
English 
Spanish 
4 min of reading 
0 comments 
DeepSeek’s new generation of artificial intelligence combines low cost, high performance, and ultra-long context, as technological tensions between China and the United States intensify
Chinese startup DeepSeek announced on Friday (24) the launch of its DeepSeek-V4 model, highlighting drastically lower costs and high performance. At the same time, the advancement rekindles the technological dispute between China and the United States, according to recent statements from the White House.
Furthermore, the company, headquartered in Hangzhou, strengthens its global presence after the impact caused in January 2025, when it presented the R1 model. At the time, the system surprised by rivaling American solutions, according to DeepSeek itself.
V4 Model Expands Capacity and Reduces Costs
The DeepSeek-V4 brings an ultra-long context, with support for up to one million tokens, according to a statement released on the WeChat platform. In this way, the model achieves capacity similar to Google’s Gemini.
-
Cherry iPhone takes shape with the Dark Cherry tone on the iPhone 18 Pro and sparks curiosity, while discreet design adjustments and a foldable iPhone indicate a bigger change happening behind the scenes.
-
A monstrous cold air mass will sweep half of Brazil in seven days with sub-zero temperatures, frosts from Rio Grande do Sul to Paraná, a historic cold snap in the Amazon, and a real possibility of snow falling in the South of the country for the first time in 2026.
-
Students from São Gonçalo surprise Brazil by securing a spot in an international robotics competition with extreme dedication and an innovative project
-
Internal link resolution test
At the same time, the company states that the system leads globally in efficiency. In other words, there is a significant reduction in computing and memory costs, according to an official post on X.
Additionally, the model was launched in two versions:
- DeepSeek-V4-Pro, with greater capacity
- DeepSeek-V4-Flash, with lower cost and greater efficiency
Meanwhile, the V4-Pro features 1.6 trillion parameters, and the V4-Flash has 284 billion, which directly influences AI decision-making.
On the other hand, in world knowledge tests, the V4-Pro was only behind the latest version of Gemini, according to DeepSeek.
Technical Impact Seen as Inflection Point
In this regard, experts point out that the launch marks a significant shift in the sector. According to Zhang Yi, founder of iiMedia, it is an “inflection point” in the AI industry.
He also highlighted that old problems, such as high cost and low efficiency in long contexts, have been overcome. Consequently, applications previously restricted to laboratories are expected to reach the market.
Therefore, commercial use tends to grow. Thus, AI tools will be able to process large volumes of text more easily and at a lower cost.
Advancement Strengthens Chinese Industry
At the same time, analyst Max Liu, a sector veteran, stated that the launch represents a milestone for Chinese artificial intelligence companies.
According to him, the advancement can increase the competitiveness of the domestic market. Furthermore, new products are expected to emerge, boosting local innovation.
Meanwhile, DeepSeek maintains an open-source strategy, allowing public access to its systems. In this way, it differentiates itself from competitors like OpenAI, which operate with proprietary models.
“DeepSeek Shock” and Privacy Concerns
Previously, the so-called “DeepSeek shock”, which occurred in 2025, provoked a strong market reaction. At the time, AI company stocks retreated, and strategies were revised.
Furthermore, the episode was compared to a “Sputnik moment”, indicating a global strategic impact. The company’s chatbot showed performance similar to ChatGPT, but with lower computational usage.
On the other hand, relevant concerns arose. The system was criticized for limitations on sensitive topics, such as historical events of 1989 in China.
Nevertheless, the use of the technology grew rapidly. Currently, DeepSeek solutions are adopted by:
- Chinese municipal governments
- healthcare institutions
- financial sector
- private companies
US accusations heighten geopolitical tension
In parallel, the White House intensified criticism of China. On Thursday (23), advisor Michael Kratsios stated that there is evidence of attempts to “steal” AI technology.
According to him, Chinese entities are conducting industrial-scale distillation campaigns. This practice, although common, was pointed out as a strategy to replicate American models.
Furthermore, the statements occurred before a possible summit between Donald Trump and Xi Jinping, scheduled for next month in Beijing.
Global scenario includes cuts and AI investments
Meanwhile, Western companies are also undergoing changes. Meta announced plans to reduce its workforce by about 10%, seeking efficiency.
At the same time, the company continues to invest heavily in artificial intelligence. Similarly, there are reports that Microsoft is evaluating staff cuts.
Therefore, the current scenario combines cost reduction, intense competition, and accelerated technological advancements.
Given this, with cheaper and more powerful models emerging rapidly, could global leadership in artificial intelligence shift in the coming years?
Be the first to react!