Measured by who actually uses what.
OpenAI's coding agent
46 stacks
#10 in AI
No reasons written yet
The frontier model out of Beijing
22 stacks
#29 in AI
3 people stack both — it isn't always either-or
Best AI tools·Judge the deck