惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Jina AI
Jina AI
博客园 - Franky
Apple Machine Learning Research
Apple Machine Learning Research
酷 壳 – CoolShell
酷 壳 – CoolShell
阮一峰的网络日志
阮一峰的网络日志
量子位
雷峰网
雷峰网
宝玉的分享
宝玉的分享
V
Visual Studio Blog
博客园_首页
小众软件
小众软件
The Cloudflare Blog
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
大猫的无限游戏
大猫的无限游戏
博客园 - 聂微东
S
SegmentFault 最新的问题
博客园 - 【当耐特】
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
博客园 - 叶小钗
月光博客
月光博客
博客园 - 三生石上(FineUI控件)
人人都是产品经理
人人都是产品经理
WordPress大学
WordPress大学

ABC News: Technology

States across the wildfire-prone Western US are using AI for early detection Shares of eBay take off on a $56 billion buyout bid from GameStop's Ryan Cohen New Mexico seeks child safety restrictions on Meta apps and algorithms in trial's 2nd phase What to Stream: 'The Drama,' MUNA, Rachel McAdams, Dan Stevens and 'The Other Bennet Sister' A citizen campaign returns iconic kiwi birds to New Zealand's capital after a century-long absence Wreckage of Coast Guard ship lost during WWI found off coast of England Apple beats out earnings estimates with continued iPhone momentum Elon Musk spars with OpenAI attorney in trial over company's evolution from a nonprofit Inside 'Scientology speedruns,' the viral trend prompting the church to bolster security Ways people are putting AI to work, from grading papers to decoding jargon Roblox to require facial scans for children under 16 in Indonesia due to new social media rules Teens embrace social media and influencers for news but remain skeptical Experts warn of rising lead risks in Africa’s solar energy boom Alphabet's first-quarter profit soars as Google's big AI bets help push stock to new highs Amazon reports increased 1Q profits and net sales fueled by cloud computing demand Meta beats revenue expectations, boosts capital spending forecast for 2026 One of America’s oldest weather observatories shows people the science behind our climate Beijing clamps down on drones: Sales banned citywide from May 1 Rare earth mining is poisoning Mekong River tributaries, threatening 'the world's kitchen' Photos show how toxic runoff from rare earth mines are risking Southeast Asia's rivers Amazon touts a 'major expansion' with OpenAI as Microsoft ties loosen Archaeologists at Pompeii use artificial intelligence to reveal face of one victim What to Stream: 'Wuthering Heights,' Kacey Musgraves, Tori Amos and a double dose of Matthew Rhys The threat of light pollution puts the world’s darkest skies in the Atacama Desert at risk Bank robber's cellphone gave him away; now Supreme Court hears his case Nation's first state moratorium on data centers vetoed by Maine's governor AI smart glasses will help visually impaired runners take on the London Marathon At Beijing auto show, Chinese carmakers flaunt new technologies Czech power company ČEZ signs deal with Rolls-Royce SMR to prepare for first small nuclear reactor Q&A: Apollo astronaut Schmitt talks about getting back to the moon and life in the universe
China's DeepSeek rolls out a long-anticipated update of i...
2026-04-24 · via ABC News: Technology

HONG KONG -- DeepSeek, the Chinese artificial intelligence startup that shook world markets last year, launched preview versions of its latest major update Friday as the AI rivalry between China and the U.S. heats up.

DeepSeek’s V4 has been keenly anticipated by users keen to test how it compares to U.S. competitors like OpenAI’s ChatGPT, Anthropic’s Claude and Google’s Gemini. Anthropic and OpenAI have accused DeepSeek of unfairly building its technology off their own.

Some industry analysts had expected the new model to arrive more than a month earlier at the start of the Lunar New Year.

DeepSeek says the new V4 open-source models, which include “pro” and “flash” versions, have big improvements in knowledge, reasoning and in their “agentic” capabilities – the ability to perform complex tasks and workflows autonomously.

V4 is a successor to V3, an AI model that DeepSeek released in late 2024.

But it was DeepSeek’s specialized “reasoning” AI model, called R1, that took markets by surprise with its release in January 2025. DeepSeek claimed it was more cost-effective than OpenAI’s similar model and it became a symbol of how China was catching up with the U.S. in technological advancements.

DeepSeek said the “V4 Pro Max” version has “superior performance” in terms of standard reasoning benchmarks relative to OpenAI’s GPT-5.2 model and Google’s Gemini 3.0-Pro. It falls “marginally” short of GPT-5.4 and Gemini 3.1-Pro, it said.

In terms of “agentic” capabilities, the Chinese company said the V4 “pro” version could outperform Claude’s Sonnet 4.5 and approaches the level of Claude's Opus 4.5 model based on its own evaluation.

The “flash” version of V4 performs on a par with the “pro” version on simple agent tasks and has reasoning capabilities closely approaching it, DeepSeek said.

“Based on the benchmark results, it does appear DeepSeek V4 is going to be very competitive against its U.S. rivals,” said Lian Jye Su, chief analyst at the technology research and advisory group Omdia.

Marina Zhang, an associate professor at the University of Technology Sydney, said DeepSeek's V4 rollout is as a “pivotal milestone for China’s AI industry”, especially as global competition intensifies in the pursuit of self-reliance in critical technologies.

Popular Reads

DeepSeek offers a free‑to‑use web and mobile chatbot. Unlike the top models from Anthropic, Google and OpenAI, it describes its technology as “open source” in the way that it enables developers access to modify and build on its core technology.

Both the V4's “pro” and “flash” versions have a 1 million token context window, a parameter of how much information an AI model can process and recall, and run on a more efficient basis, the startup said. That is a significant improvement from before, since the V3 supported a 128,000 token context window.

A report from Microsoft in January showed use of DeepSeek has been gaining ground in many developing nations.

However, some analysts remain skeptical. Ivan Su, a senior equity analyst at Morningstar, said while V4 is a “competent” follow-up, it’s not as big a breakthrough as the rollout of R1.

“Domestic competition has intensified significantly since R1’s release,” Su said. “Against U.S. models, DeepSeek’s own evaluation suggests its capabilities largely match on most fronts, but independent evaluations are needed before final conclusions can be drawn.”

In February, Anthropic accused DeepSeek and two other China-based AI laboratories of “industrial-scale campaigns” to “illicitly extract Claude’s capabilities to improve their own models.” It said they did that using a technique called distillation that “involves training a less capable model on the outputs of a stronger one.” OpenAI made similar allegations in a letter to U.S. lawmakers.

This week, Michael Kratsios, chief science and technology adviser to U.S. President Donald Trump, also accused foreign tech companies “principally based in China” of distilling leading U.S. AI systems and “exploiting American expertise and innovation.”

China’s embassy in Washington hit back at the allegations, describing them as “unjustified suppression of Chinese companies by the U.S.”

___

O’Brien reported from Providence, Rhode Island.