09 How Good Are Chinese Models' Programming Capabilities Now
Recently tested 2 Chinese models, one is glm5, another is qwen3.5 plus
In 2025 Chinese models couldn't be used to write code, or could barely make do, hand-holding was very hard to use
Result this historical turning point happened during Spring Festival two days ago
In the past year my requirements for model programming capabilities also changed greatly, I defined 2 stages:
- AI Programming 1.0 (2023~before December 2025)
In this stage, AI that can write code needs to meet 2 requirements:
- Can understand requirements: understand what development task is about
Purchase required to continue
This is a paid article. After signing in, your purchase will be unlocked automatically.
No comments yet. Be the first to share your thoughts.