The current balance of power in open models
A report finds Chinese open-weight AI models now outperform American counterparts in downloads and benchmarks.
A report briefing Congressional members on the state of open-weight models in U.S.-China competition. Chinese labs lead in open-weight model performance and downloads, with models like GLM-5.3 and Kimi K3 crossing key capability thresholds. American open-weight models lag behind both Chinese models and the closed American frontier.
- Chinese open-weight models lead in downloads and benchmarks, surpassing American counterparts.
- Z.ai's GLM-5.3 and Moonshot AI's Kimi K3 show step-change improvements in agentic capabilities.
- American open-weight models lag 6-9 months behind closed frontier models, while Chinese models are 2-5 months behind.
- The report was prepared for a Congressional briefing on U.S.-China AI competition.
- Products
- ChatGPTClaude Code
- Countries
- United StatesChina
From the original article
13 min read at interconnects.ai
I was recently invited to brief a group of Congressional members and staff on the state of open-weight models in the lens of U. S. -China competition.
By Nathan Lambert · interconnects.ai
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.