Qwen 3 is here! 🎉 In this video, I dive into Alibaba’s latest series of models, featuring six dense models ranging from 600 million to 32 billion parameters and two mixture of experts models. The flagship model boasts 235 billion parameters with 22 billion active ones. 🚀 I cover how to access and utilize these models, including platforms like Hugging Face and Kaggle. These models excel at coding tasks, outperforming others like Gemini 2.5 Pro. 💻 I’ll also touch on some unique features like hybrid thinking mode and agentic capabilities. For more details, you can pull these models from Hugging Face or run them locally. Don’t miss out on trying them at chat.qwen.ai. Enjoy!
00:00 Introduction to Qwen Three
00:11 Overview of New Models
00:44 Performance Highlights
02:18 Accessing the Models
02:40 Hybrid Thinking Model
04:15 Training and Data Details
05:12 Comparison with Other Models
06:54 User Impressions and Reactions
08:46 Conclusion and Final Thoughts
source
