DeepSeek V3 Upgrade Challenges AI Giants with Open-Source Efficiency
16 Sources
[1]
DeepSeek V3 Is Now Reportedly the Best Non-Reasoning AI Model
DeepSeek V3 is now the best non-reasoning AI model on the market, besting OpenAI, Grok and Google, according to firm Artificial Analysis on Tuesday. While AI companies trade blows for the top spot as new updates release, what makes DeepSeek V3's accomplishment noteworthy is that it's the first
[2]
DeepSeek's V3 AI model gets a major upgrade - here's what's new
Not that it ever left, but it appears Chinese AI startup DeepSeek is back in the news -- this time with an updated version of its V3 model, released in December. Also: The best AI chatbots On Tuesday, the company officially announced V3-0324, named after its release month and day. A day earlier,
[3]
Chinese AI start-ups overhaul business models after DeepSeek's success
Chinese artificial intelligence start-ups are overhauling their business models as they fight to remain competitive following the widespread adoption of rival DeepSeek's technology across the country. Zhipu, once considered China's most prominent large language model start-up, has pinned its hopes
[4]
China's DeepSeek releases AI model upgrade, intensifies rivalry with OpenAI
BEIJING, March 25 (Reuters) - Chinese artificial intelligence startup DeepSeek released a major upgrade to its V3 large language model, intensifying competition with U.S. tech leaders like OpenAI and Anthropic. The new model, DeepSeek-V3-0324, was made available through AI development platform
[5]
DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that's a nightmare for OpenAI
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Chinese AI startup DeepSeek has quietly released a new large language model that's already sending ripples through the artificial intelligence industry -- not just for its
[6]
Deepseek's new AI is smarter, faster, cheaper, and a real rival to OpenAI's models
DeepSeek claims its AI models can match or beat those of American AI developers like OpenAI and Anthropic DeepSeek dropped a major upgrade to its AI model this week, which has people buzzing almost as much as they did when the Chinese AI startup first made its splash earlier this year. The new
[7]
DeepSeek in China-US AI competition
DeepSeek influences US-China competition and cooperation, temporarily impacting the US AI industry while also triggering stricter chip controls on China. DIGITIMES observed that the rise of DeepSeek has influenced the AI development directions in both China and the US, bringing significant changes
[8]
DeepSeek-V3 is the Highest Scoring Non-Reasoning Model - 'A Milestone for Open Source'
The model outperformed all other non-reasoning models across several benchmarks but trailed behind DeepSeek-R1, OpenAI's o1, o3-mini, and other reasoning models. DeepSeek on Monday announced a new update to its general-purpose AI model DeepSeek-V3. The updated model 'DeepSeek V3-0324' now ranks
[9]
DeepSeek's new open-source colossus upends the AI status quo
Just two days ago, Chinese AI startup DeepSeek quietly dropped a bombshell on Hugging Face: a 685-billion-parameter large language model called DeepSeek-V3-0324. While some innovations arrive with fanfare, this release was different. No splashy press briefings. No polished blog posts. Just a
[10]
DeepSeek's V3 upgrade challenges OpenAI and Anthropic in global AI race
Alibaba has unveiled its latest AI reasoning model, QwQ-32B, whose performance is on par with DeepSeek's R1 model despite using significantly fewer parameters. Chinese artificial intelligence startup DeepSeek released a major upgrade to its V3 large language model, intensifying competition with
[11]
China's DeepSeek Unveils Latest Update in Race With OpenAI
DeepSeek stunned the industry and sparked a selloff in US markets DeepSeek released updates to its V3 model that promise to deliver better programming capabilities, underscoring the Chinese AI startup's intent to remain a step ahead of competitors. The V3-0324 update -- posted on Hugging Face
[12]
China's DeepSeek releases AI model upgrade, intensifies rivalry with OpenAI
The latest model demonstrates significant improvements in areas such as reasoning and coding capabilities compared to its predecessor, with benchmark tests showing enhanced performance across multiple technical metrics published on Hugging Face.Chinese artificial intelligence startup DeepSeek
[13]
China's DeepSeek Ups The Heat In OpenAI Rivalry, Upgrades V3 Model Improving Coding And Reasoning Capability - Alphabet (NASDAQ:GOOG), Amazon.com (NASDAQ:AMZN)
How to Spot the Market Bottom: Matt Maley has navigated every major market turn in the last 35 years, and on Wednesday, March 26, at 6 PM ET, he's revealing how to recognize when the worst is over, the trades to make before the next bull market takes off, and the stocks and sectors that will lead
[14]
China's DeepSeek reveales updated AI model as OpenAI rivalry intensifies - Reuters By Investing.com
Investing.com - DeepSeek has unveiled an update to its low-cost artificial intelligence model, intensifying the Chinese start-up's bid to rival AI industry leaders like ChatGPT-maker OpenAI and Anthropic, Reuters reported on Tuesday. The newest version of the offering, DeepSeek-V3-0324, has become
[15]
DeepSeek V3-0324 update advances AI accessibility and performance By Investing.com
Investing.com -- The Chinese AI research group, DeepSeek, released the latest update to its DeepSeek V3 model, named DeepSeek V3-0324, on March 24, 2025. This open-source model presents significant enhancements in reasoning, coding, and frontend development capabilities. It surpasses its
[16]
China's DeepSeek releases AI model upgrade, intensifies rivalry with OpenAI
BEIJING (Reuters) - Chinese artificial intelligence startup DeepSeek released a major upgrade to its V3 large language model, intensifying competition with U.S. tech leaders like OpenAI and Anthropic. The new model, DeepSeek-V3-0324, was made available through AI development platform Hugging Face,
Share
Copy Link
Chinese AI startup DeepSeek releases a major upgrade to its V3 language model, showcasing improved performance and efficiency. The open-source model challenges industry leaders with its ability to run on consumer hardware.
DeepSeek V3 Upgrade Challenges AI Industry Leaders
Chinese AI startup DeepSeek has released a significant upgrade to its V3 large language model, dubbed DeepSeek-V3-0324, intensifying competition with industry giants like OpenAI and Anthropic. The new model, which appeared on AI repository Hugging Face with little fanfare, demonstrates substantial improvements in reasoning and coding capabilities compared to its predecessor
1
.Breakthrough Performance and Efficiency
DeepSeek-V3-0324 employs a mixture-of-experts (MoE) architecture, activating only about 37 billion of its 685 billion parameters during specific tasks. This selective activation represents a paradigm shift in model efficiency, allowing performance comparable to much larger fully-activated models while drastically reducing computational demands
5
.The model incorporates two additional breakthrough technologies:
- Multi-Head Latent Attention (MLA): Enhances the model's ability to maintain context across long passages of text.
- Multi-Token Prediction (MTP): Generates multiple tokens per step, boosting output speed by nearly 80%
5
.
Consumer Hardware Compatibility
One of the most striking features of DeepSeek-V3-0324 is its ability to run on consumer-grade hardware. Early reports suggest that a 4-bit quantized version of the model can achieve speeds of over 20 tokens per second on an Apple Mac Studio with an M3 Ultra chip
5
. This development challenges traditional assumptions about the infrastructure requirements for top-tier AI model performance.Open-Source Strategy and Market Impact
DeepSeek's decision to release the model under an MIT license, making it freely available for commercial use, exemplifies a growing trend among Chinese AI companies
2
. This open-source approach contrasts sharply with the closed, API-centric strategies of Western leaders like OpenAI and Anthropic5
.The availability of cutting-edge models like DeepSeek-V3-0324 is transforming China's AI ecosystem, enabling startups, researchers, and developers to build upon sophisticated AI technology without massive capital expenditure
3
.Related Stories
Competitive Landscape and Industry Shifts
DeepSeek's rapid ascent has prompted other Chinese AI startups to reevaluate their strategies:
- Zhipu is focusing on enterprise sales and considering an IPO.
- 01.AI has stopped pre-training large language models to concentrate on selling tailored AI business solutions using DeepSeek's models.
- Baichuan is narrowing its focus to the healthcare market.
- Moonshot has reduced marketing for its Kimi chatbot to prioritize model training
3
.
Performance Claims and Future Implications
Early testers report that DeepSeek-V3-0324 may now be the best non-reasoning AI model, potentially surpassing Claude Sonnet 3 from Anthropic
5
. If validated, this claim would solidify DeepSeek's position as a formidable competitor in the global AI market.As DeepSeek and other Chinese AI companies continue to innovate and release open-source models, the competitive landscape of the AI industry may see significant shifts, challenging the dominance of established Western players and potentially accelerating the pace of AI development worldwide.
References
Summarized by
Navi
Related Stories
Recent Highlights
1
OpenAI's AI Math Breakthrough on Millennium Problem Sparks Fierce Attribution Battle
Science and Research
2
Trump Dismisses AI Regulation Calls, Targets Anthropic CEO Amid Growing Safety Concerns
Policy and Regulation
3
Nvidia CEO Jensen Huang Declares AGI Has Arrived With OpenAI's GPT-6 Astra Release
Technology
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
