China's Open AI Models Now Have Twice the Downloads of America's. Cursor and Perplexity Already Build on Them.

Nathan Lambert prepared a briefing for members of Congress on where open-weight AI models stand today. Open-weight models are AI systems whose underlying files get published for anyone to download and run on their own hardware. That’s different from a closed model, which you can only reach through a company’s paid app or API, a set of instructions a program uses to talk to a service over the internet. His conclusion: Chinese labs have not just caught up. They have pulled ahead, and American companies are already quietly building on their models.

What he said

Lambert writes that the shift already happened, not that it’s coming. “Chinese AI companies are the clear leaders in open weight models,” he states. He puts a date on when that lead opened up: “Chinese open-weight models surpassed American open-weight models in these two key areas about 18 months ago.”

He backs that with download and usage data. “China’s download lead has grown to about 1.6B, with a total of 3.2B downloads, twice that of America’s total,” he writes. On where developers are actually sending API traffic today, he points to OpenRouter, a marketplace that routes requests across many AI providers. Its numbers show which models get used in real products, not just which ones get press coverage. “Chinese models have grown from ~70% market share to over 80% of usage,” he writes.

The adoption claim is the one that should get builders’ attention. “Many prominent technology companies and startups have been building on Chinese open-weight models for their AI features, such as Harvey, the legal agent, Cursor, the coding agent, and DoorDash’s use of Kimi models, Airbnb’s use of Qwen, or Perplexity’s use of DeepSeek,” Lambert writes.

Who he is and why the take carries weight

Lambert writes Interconnects, an independent newsletter that tracks AI model releases, training methods, and industry strategy. Before that, he spent two and a half years at the Allen Institute for AI (Ai2). He led post-training research there, the work of fine-tuning an already-built model’s behavior after its initial training, and built the Olmo and Tulu open model families. He left this past June to work independently. He’s also the author of a book on reinforcement learning from human feedback, the training technique behind how today’s chatbots learn to follow instructions.

This piece is adapted from remarks he prepared for a congressional briefing, not a blog post written to get clicks. That raises the bar: the numbers need to hold up under questioning from staff who will fact-check them. Lambert also defines his terms carefully. Open-weight models are ones where a company publishes the trained files under a license but keeps its training data and code private. Fully open-source models publish the training data and code too. Most of what gets called “open” today, including the Chinese models he’s describing, is open-weight, not open-source.

What he gets right, and where it’s incomplete

The download and usage numbers are Lambert’s own count from public platforms. They point in one direction clearly enough that the trend itself isn’t really in dispute. The adoption list, naming Cursor, Perplexity, DoorDash, and Airbnb by name, is the strongest part of the piece. It shows the shift isn’t confined to hobbyists comparing benchmark scores. It’s already inside products people use.

Where the piece is thinner is on causes and endpoint. Lambert names capacity gaps, regulatory posture, and talent distribution as contributing factors, but doesn’t rank which one matters most. His benchmark comparison also has a limit worth naming. He cites Chinese models Z.ai’s GLM-5.3 and Moonshot AI’s Kimi K3 scoring higher than the American open models he names. That comparison uses the Artificial Analysis Intelligence Index, a third-party benchmark that combines several tests into one capability score. It measures open models against other open models. It doesn’t measure them against the closed, API-only models from OpenAI or Anthropic that most consumer AI products still run on. His closing recommendation, that the US should “continue to enable investment in open models,” is his own policy prescription for Congress. Treat it as his position, not a documented outcome.

Why it’s notable

Most coverage of the US-China AI race focuses on frontier, closed models: which lab has the smartest chatbot. Lambert’s briefing is about a different, more practical layer. It’s about which models are cheap and available enough for a startup to build on, run on its own servers instead of paying a vendor for cloud access, or fine-tune for a narrow task. That’s the layer where cost and control matter most for builders. By his numbers, China has taken a clear lead there.

What it means for builders

Cost and licensing terms decide the model choice more often than benchmark scores do. Lambert’s numbers say some of the best-performing, cheapest-to-self-host open models today come from Chinese labs like Z.ai, Moonshot AI, and Alibaba’s Qwen team. Companies like Cursor and Perplexity are already treating that as a normal engineering decision, not a controversial one.

That doesn’t mean every builder should switch. It means the choice deserves the same evaluation you’d give an American vendor. Check the license terms for commercial use. Test the model against your actual task instead of a general benchmark. Don’t rule a model out purely because of where the lab is based. The gap Lambert describes is a build-cost decision now, not just a policy debate.

End of article