Ali launched a new inference model with only 1/21 parameters comparable to DeepSeek R1
2025-03-05 22:51:07
On March 6th, Ali Qwen team officially released their latest research results - QwQ-32B large language model, QwQ-32B achieved a performance leap with reinforcement learning under the condition of only about 1/21 of the parameters of DeepSeek-R1. < br > < img src = "https://flash-scdn.jin10.com/fae1ace9-d810-4fe7-9eeb-9a6bbefb8a18.png" referrerpolicy = "no-referrer" >
Disclaimer:
1. The information provided does not constitute investment advice. Investors should make independent decisions and bear all risks themselves.
2. The copyright of this content belongs to the original author. The views expressed herein are solely those of the author and do not represent the stance or position of this website.
