Subscribe to Our Newsletter

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks
Alibaba says its Qwen3 models beat DeepSeek R1 and others
Photo by Igor Omilaev / Unsplash

Alibaba says its Qwen3 models beat DeepSeek R1 and others

But what might set Qwen3 apart isn’t just its scale. It's accessibility. 

Ogbonda Chivumnovu profile image
by Ogbonda Chivumnovu

The race to lead in AI has never been tighter, and the Chinese tech company, Alibaba is making bold moves to stay in the game. On Tuesday, the tech giant unveiled its latest Qwen3 model family, a third-generation series of open-source AI models built to rival top global competitors like OpenAI and Google.

The Qwen3 lineup includes eight models ranging from 600 million to a massive 235 billion parameters. In the world of AI, more parameters often signal more intelligence, and Alibaba claims these models outperform many domestic and international rivals, including OpenAI’s o1 model, Google’s Gemini and DeepSeek's R1, on tasks like coding, math, and complex reasoning.

What might set Qwen3 apart isn’t just its scale, but its accessibility. Alibaba says the models are open source and claims they've already powered over 100,000 derivative models. If accurate, that could make Qwen one of the largest open-source AI ecosystems currently available, potentially even rivalling Meta’s Llama community, which has dominated much of the open-source conversation until now.

The models are trained on 36 trillion tokens and reportedly support 119 languages and dialects, three times the coverage of Qwen2.5. This expanded multilingual reach could position Qwen3 as a strong contender in non-English markets, where language support has often been limited or inconsistent.

OpenAI rolls out o3-mini to rival DeepSeek’s R1
More than just an incremental upgrade, o3-mini is a clear signal to DeepSeek that OpenAI isn’t ceding ground easily.

Qwen3 is now integrated into Alibaba’s chatbot and available on GitHub, Hugging Face, and the company’s ModelScope platform, making it easy for developers and researchers to experiment and build.

This launch lands at a time when AI model releases are accelerating. On the global front, OpenAI recently introduced its o3 and o4-mini reasoning models, while Meta unveiled Llama 4 Scout and Maverick, multimodal systems built for video, audio, and image processing, with the Behemoth model still in preview. Locally, Baidu has released two new models, and DeepSeek’s R2 is reportedly on the horizon.

Alibaba’s move isn’t just about matching capabilities; it’s about leadership in the open-source AI space. And with China's growing appetite for homegrown tech solutions, Qwen3 could position Alibaba as a key player not just in China, but in the global AI race.

As the field edges closer to artificial general intelligence, tools like Qwen3 show how fast the bar is rising and how fiercely companies are competing to define the future of intelligent systems.

Ogbonda Chivumnovu profile image
by Ogbonda Chivumnovu

Subscribe to Techloy.com

Get the latest information about companies, products, careers, and funding in the technology industry across emerging markets globally.

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks

Read More