DeepSeek's updated R1 AI model equals coding ability of Google, Anthropic in new benchmark
The latest model update from Chinese
artificial intelligence (AI) start-up
DeepSeek has matched the coding performance of industry heavyweights
Google and Anthropic, according to the latest results from WebDev Arena, a real-time AI coding competition.
The updated version of DeepSeek-R1 tied for first place with Google's Gemini-2.5 and Anthropic's Claude Opus 4 on the WebDev Arena leaderboard, which evaluates large language models (LLMs) on their ability to solve coding tasks quickly and accurately. The Hangzhou-based company's R1 scored 1,408.84, in line with Opus 4's 1,405.51 and Gemini-2.5's 1,433.16.
The quality of the models' output is evaluated by humans, who determine the scores. DeepSeek's reasoning model has consistently performed at levels close to leading models in various benchmark tests since it was unveiled in January, despite significantly lower training costs.
DeepSeek quietly updated R1 in late May, marking its first revision since its high-profile debut. The start-up released R1-0528 on the open-source AI developer community Hugging Face, calling it a 'minor upgrade' and offering no details on the changes. It later said the updated model had improved in reasoning and creative writing capabilities, with a 50 per cent reduction in hallucinations – instances where AI generates misleading information with little factual basis.
The R1 update attracted attention from the developer community amid widespread anticipation for DeepSeek's next-generation reasoning model, R2. The company has said little about when it might release its big follow-up.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
South China Morning Post
30 minutes ago
- South China Morning Post
China's possible Israel-Iran peace role, robot firms' big salaries: SCMP daily highlights
Catch up on some of SCMP's biggest China stories of the day. If you would like to see more of our reporting, please consider subscribing China is expected to play an 'active' role in trying to broker a ceasefire between Israel and Iran, but analysts warned there may be limits to what it could achieve. A week after the second round of high-profile trade negotiations between Beijing and Washington, analysts break down what happened before, during and after. The number of Chinese suppliers attending the Paris Air Show has more than doubled for this year's edition. Photo: Xinhua China's presence at the Paris Air Show – the globally renowned civil aviation expo – is usually reduced to Beijing's biggest names in the field. In particular, conversation tends to focus on the Commercial Aircraft Corporation of China (Comac), the maker of the C919 passenger jet. But things have changed this year. Dozens of lesser-known Chinese firms from across the supply chain have flocked to the European capital to showcase their products to Western buyers.
South China Morning Post
an hour ago
- South China Morning Post
Reaping of the sow: China to reduce pig count by 1 million amid low prices, deflation risk
China's national breeding sow inventory will be reduced by 1 million from the current level of 40.38 million to ease an oversupply of pork in the market that has suppressed swine prices and raised deflationary pressures in the economy. While specifics of the reduction were limited, it would take the national sow herd size down to 39.5 million, said in an exclusive report, citing a plan proposed by the Ministry of Agriculture and Rural Affairs last week. The ministry was looking to ease industry losses caused by an oversupply of hogs and persistently low pork prices, the website, which is an online tech and financial news platform, reported on Tuesday. In addition to cutting the number of sows, regulators have reportedly introduced stricter rules for pig farms, like prohibiting pigs that have already reached the slaughtering standard from continuing to be fed to increase their weight before being sold – an industry practice blamed for worsening short-term oversupply and further depressing prices. The measures are not only aimed at restoring a healthier supply-demand balance in the pork sector, but also at easing deflationary pressure in the broader economy, since the price of pork is highly weighted in China's consumer price index that tracks the price changes of a basket of goods and services purchased by consumers. 2025 could be another year with persistent deflationary pressures, unless the stimulus is big enough to create another credit upcycle, according to a report by Macquarie last week. China's GDP deflator has fallen for eight quarters in a row, marking the longest deflationary streak in the past four decades, the report said, referring to the measurement of the overall price level for new, domestically produced goods and services – making it a broad measure for inflation.
South China Morning Post
an hour ago
- South China Morning Post
Hong Kong could serve as stablecoin test bed amid China's effort to raise yuan's profile
Hong Kong could serve as a stablecoin test bed to boost the internationalisation of the yuan as Beijing puts more focus on the digital version of its currency, according to the chairman and CEO of HashKey Group. Advertisement 'Due to its 'one country, two systems' characteristics, Hong Kong's stablecoins can serve as a testing ground for the mainland, providing both experience and lessons,' said Xiao Feng, who heads one of the city's licensed cryptocurrency exchange operators, on Wednesday. Earlier in the day, People's Bank of China (PBOC) governor Pan Gongsheng spoke about stablecoins – digital tokens pegged to a reference asset like a fiat currency – at the high-profile Lujiazui Forum. Pan said that emerging technologies such as blockchain and distributed ledgers were rapidly driving the development of central bank digital currencies (CBDCs) and stablecoins, reshaping traditional payment systems and significantly shortening cross-border payment chains while posing 'significant challenges to financial regulation'. HashKey Group CEO Xiao Feng. Photo: Handout Pan added that innovations like smart contracts and decentralised finance would continue to advance the evolution of cross-border payment systems. Advertisement HashKey's Xiao viewed the speech as a 'positive sign' and that stablecoins could be on Beijing's radar.
