🇨🇳🇺🇸Chinese AI models are up to 50 times cheaper than their American counterparts on a per-token basis.
Rohan Paul Twitter · Rohan Paul (@rohanpaul_ai) · 2026-06-27
A J.P. Morgan report finds Chinese AI models including Qwen, DeepSeek, and Kimi are up to 50 times cheaper per token than American counterparts, with Chinese firms capturing over 45% of OpenRouter traffic by April 2026 and enterprises shifting to model routing to control costs.
Appears in
Extraction
Topics: ai-pricingchinese-ai-modelsenterprise-aiai-market-dynamicssemiconductor-industry
Claims
- Chinese AI models are up to 50 times cheaper per token than American equivalents, with Qwen, DeepSeek, and Kimi creating direct pricing pressure on OpenAI and Anthropic.
- Chinese firms accounted for over 45% of all traffic on OpenRouter by April 2026, up from under 2% in late 2024.
- Enterprise AI tokens may become commoditized as many business tasks can run on smaller open models rather than frontier models.
- AI has driven 65–80% of S&P 500 returns, profits, and capex since ChatGPT, which J.P. Morgan characterizes as showing investor overexcitement in semiconductors.
- Custom AI chips from Google, Amazon, Microsoft, and Meta are gaining share from NVIDIA by potentially cutting total compute cost by 30–40%.
Key quotes
Chinese AI models are up to 50 times cheaper than their American counterparts on a per-token basis.
Chinese firms accounted for over 45% of all traffic on the AI aggregation platform OpenRouter by April 2026, up from under 2% in late 2024.
Companies are not abandoning AI, they are using model routing, which sends easy tasks to cheaper models and saves premium models for hard reasoning, code, and long-context work.