Web programming crowdsourced testing ranking: DeepSeek-R1 surpasses Claude 4 to crown the world's first.

Web Programming Public Testing Ranking: DeepSeek-R1 Surpasses Claude 4 to Crown as the World's First

via cnBeta.COM - Chinese Industry Information Station (author: Source: QbitAI)

Telegraph
Web Programming Public Testing Ranking: DeepSeek-R1 Surpasses Claude 4 to Crown as the World's First

Is the throne of programming king Claude shaky?? The latest battle report from the large model arena is out. The new version of DeepSeek, R1, has won the first place in web programming, narrowly beating Claude Opus 4. You know, Claude Opus 4 is recognized as the "world's strongest coding model". So, how on earth did DeepSeek-R1-0528 manage to defeat Claude Opus 4 in programming? Judging from the name, you might think it's a minor version update, but in fact - it's almost on par with OpenAI o3-high on LiveCodeBench, and many netizens even speculate that it's the legendary R2. Looking at it this way, in terms of programming, neither side seems to be easy to mess with~ So, without further ado...