For iOS, Deepsek has surpassed Openai’s chat to top the “Top Free Apps” charts of App Store in US. The nominee Chinese company released the Open-SOS Dipsec-R1 Artificial Intelligence (AI) model last week, which improves the O1A AI model of openiI in several benchmarks. This release has made the company and its AI Chatbot a matter of the city, with many Silicon Valley Tech leaders reacted to its sudden growth. In particular, the Dipsek app is free to fully use, in which no membership has been declared so far.
Deepsac goes ahead of chat
Ever since Openai launched Chatgpt for the iOS app in May 2023, it has been one of the top free applications on the continuous app store and remains the highest ranked AI app on the platform. However, this throne was recently taken by the Deepsek app. reached Top of the chart. The growth is being held responsible for the recent launch of the Deepsek-R1A AI model.
In particular, the Hangzo-based AI firm is very rarely known that was established in 2023 and many open-sources large language models (LLMs) have been released. While US-based tech companies have also released open-source models, there is a remarkable mention in the Meta community, no one comes close to the capabilities and scale of the Deepsek model. The company claims to manufacture models with a cost of significantly cheaper, $ 6 million (about 51.8 crores) compared to the AI ​​model of this scale.
Many Silicon Valley-based technical leaders have reacted to the sudden growth of these models. Venture capitalist marik andresen called R1 AI model “I have ever seen one of the most amazing and impressive successes,” while Perplexity AI co-founder and CEO Arvind Srinivas Congrats Chinese firm to become the first AI app to defeat Chatgpt.
In a linkedin PostYan Lake, AI’s Godfather and current vice -president and one of the main AI scientists in Meta, said, “Open Source Model Owners are crossing people.” Holger ZSCHAUPITZ, a senior financial reporter at the German Free-to-air television news channel Well, Argued Deepsek “may represent a large threat to American equity markets.”
Deepsek is not the only one
While Deepsek has broken the tendency of open-sourcing companies to small and less competent AI models keeping the Frontier Model under an ownership payment, it is not only one. Another Chinese firm, km AI, Announced Release of km k1.5 AI model. The company claims that it is the “O1-Level Multimodal LLM that outperforms the GPT-4o and Cloud Sonnet 3.5 on several benchmarks”.
In particular, Kimi AI has made the web version of its chatbot completely independent to use with unlimited use as Deepsek. The AI ​​model can also demonstrate real -time web discoveries, analyzes 50 files in a variety of formats, as well as in image understanding abilities. While it has a technical report Available On Github, it is not currently available in open-sources.
