Alibaba releases and opens the Tongyi Qianwen QwQ-32B inference model, with performance comparable to DeepSeek-R1

Alibaba Cloud's new open source reasoning model Tongyi Qianwen QwQ-32B, with performance comparable to DeepSeek-R1!

On March 6, Alibaba Cloud officially announced the open source of its latest research and development inference model - Tongyi Qianwen QwQ-32B. The model has achieved significant improvements in math, code and general capabilities, with performance levels comparable to the industry-leading DeepSeek-R1 model. This news stimulated Alibaba's stock price to rise sharply, once rising by more than 8%.

Alibaba Cloud confirmed in several authoritative benchmark tests that Qianwen QwQ-32B's performance far exceeds OpenAI-o1-mini and is on par with DeepSeek-R1 in overall performance. What’s even more remarkable is that while maintaining high performance, the model significantly reduces deployment costs and can even be deployed locally on consumer graphics cards.

Qianwen QwQ-32B also integrates the functions of the agent Agent, has critical thinking ability, and can adjust the reasoning process based on environmental feedback. Tongyi’s team said that in the future, the combination of agents and reinforcement learning will be further explored to achieve longer-term reasoning and ultimately strive to achieve the goal of general artificial intelligence (AGI).

It is worth noting that Alibaba Cloud adopts the loose Apache2.0 open source protocol, allowing developers around the world to download and commercialize Qianwen QwQ-32B models for free. Users can also experience the latest features of the model for free through Tongyi APP.

The above is the detailed content of Alibaba releases and opens the Tongyi Qianwen QwQ-32B inference model, with performance comparable to DeepSeek-R1. For more information, please follow other related articles on the PHP Chinese website!