Qwen 2.5 and DeepSeek R1 Keep US AI Firms in a Limbo as Stock Prices Struggle To Bounce Back

As the global tech sector was catching up to DeepSeek's R1 AI model launch last week, Alibaba (NYSE:BABA) surprised everyone by releasing the Qwen 2.5 Max large language model on the first day of the Lunar New Year.

Alibaba claimed that Qwen 2.5 Max topped DeepSeek's R1 across various benchmarks and showed impressive results compared to OpenAI's ChatGPT-4o and Meta Platform's (NASDAQ:META) Llama-3.1-405B model.

Alibaba's view that Qwen 2.5 can take on top AI models from domestic and foreign rivals has slowed down the rebound of America's "magnificent seven" stocks after the DeepSeek R1 AI model rocked the sector. Stocks like Nvidia (NASDAQ:NVDA) have tanked by over 15% in the past five days, while Elon Musk's Tesla (NASDAQ:TSLA) was down over 6.2% in the same period. The EV maker also missed its quarterly earnings expectations.

Cheap, high-performance Chinese AI alternatives are compelling global investors to question the narrative that top AI firms require billions of dollars in capital to develop leading AI models. Furthermore, the high capital spending has yet to generate comparable revenue returns. The DeepSeek R1 model was supposedly built for £4.49 million using less-advanced GPUs and relatively lower computing power.

US President Donald Trump said the impressive performance of Chinese AI models should serve as a "wake-up call" to Silicon Valley. However, he reportedly sees the likes of DeepSeek as an opportunity rather than a threat, as these AI models can help drive costs down.

Why Is Qwen 2.5 Special?

Alibaba claims its new AI model could prove to be superior to rivals across diverse metrics. The Qwen 2.5 boasts new-age capabilities for natural language processing, image and audio processing, chatbot role-play, text generation, and AI tool integration. Alibaba also ensured the AI model delivers better efficiency for faster response times and lower computational costs, which could make it a practical choice for businesses of all sizes.

Although Qwen 2.5 showcases features like long-context understanding and structure data processing, its standout feature is the multilingual support in 29 languages, including English, Chinese, Spanish, French, and Arabic, making it a valuable tool across geographies.

Furthermore, Qwen 2.5 is multimodal, which enables it to process images and audio inputs for diverse applications, ranging from creative content creation to critical data interpretation.

The AI model is trained on 18 trillion tokens to facilitate the creation of high-quality, context-aware text with high accuracy. Meanwhile, the long-context processing capabilities allow a better understanding of complex topics and more detailed responses. For better accessibility, Alibaba has integrated Qwen 2.5 into Qwen Chat so that users can easily interact with the AI model.

The latest developments in the Mainland have left top US tech firms scrambling to refine their AI models. OpenAI CEO Sam Altman said DeepSeek's AI model is impressive and that his company "will obviously deliver much better models, and also, it's legit invigorating to have a new competitor!"

The tech leader said that OpenAI will also launch "some releases" while hinting that increased compute power is integral to ensuring OpenAI continues to lead the global AI race.

Disclaimer: Our digital media content is for informational purposes only and not investment advice. Please conduct your own analysis or seek professional advice before investing. Remember, investments are subject to market risks and past performance doesn't indicate future returns.