AI News

Alibaba’s New Qwen 2.5-Max Can Outperform DeepSeek

Alibaba’s New Qwen 2.5-Max Can Outperform DeepSeek

In the last few days, one after another, numerous AI models have been introduced, making waves across the world. While tech savies are busy discussing the Chinese AI model’s impact on the world, Alibaba has released an upgraded version of its AI model, Qwen 2.5-Max, claiming that it could outperform DeepSeek-V3.

The model is now accessible through Qwen Chat, which allows users to interact with it for various tasks, similar to other generative AI models. 

Alibaba’s Qwen 2.5-Max vs DeepSeek V3

“Qwen 2.5-Max outperforms … almost across the board GPT-4o, DeepSeek-V3 and Llama-3.1-405B,” Alibaba Cloud stated on its official WeChat account. This bold statement clears that Alibaba directly wants to compete with top-notch OpenAI and Meta’s leading models. 

The release comes after DeepSeek’s AI assistant and R1 model launch, which also became a reason behind the catastrophic decline in Silicon Valley’s tech stocks. However, DeepSeek’s efficiency raised concerns about the sustainability of heavy AI investments in the West.

The meteoric rise of DeepSeek has shaken the landscape of AI innovations, not only challenging foreign rivals but also forcing domestic tech giants to accelerate their AI development. 

According to Alibaba Cloud, Qwen 2.5-Max overshadows DeepSeek V3 in different parameters, such as Arena-Hard, LiveBench, LiveCodeBench, and GPQA-Diamond, as well as MMLU-Pro.

The competition also escalated when TikTok’s parent company, ByteDance, announced an updated AI model just two days after DeepSeek-R1’s debut. During the launch, ByteDance claimed its model outperformed OpenAI’s o1 in AIME, which is a benchmark for AI instruction. 

Source: https://www.bloomberg.com/news/articles/2025-01-29/alibaba-touts-new-ai-model-superior-to-deepseek-s-and-meta-s/

Latest Stories:

OpenAI Accuses DeepSeek of Stealing AI Tech, Microsoft Probes

Gujarat CM Opens AI Center at GIFT City, Developed by Microsoft

Rajpalsinh Parmar
Rajpalsinh has been decoding the AI universe for three years, turning tech jargon into tales of wonder and possibility. With a knack for making the abstract tangible, he brings AI's potential to life for everyone.

Leave a reply

Your email address will not be published. Required fields are marked *

More in:AI News