Token Tenfold Increase in Two Months: Power Anxiety and Pricing Logic at the OpenClaw Roundtable of the Zhongguancun Forum

According to monitoring by 1M AI News, Yang Zhilin, founder of the Dark Side of the Moon, hosted a roundtable on OpenClaw and AI open source themes at the 2026 Zhongguancun Forum. Guests included Zhang Peng, CEO of Zhiyu Huazhang, Xia Lixue, co-founder and CEO of AI computing infrastructure company Wuwen Xinqiong, Luo Fuli, head of Xiaomi’s MiMo large model, and Huang Chao, assistant professor at the University of Hong Kong, covering three levels: models, computing infrastructure, and Agent applications. Xia Lixue provided an intuitive data point: since the end of January, the token usage at Wuwen Xinqiong has doubled every two weeks, totaling a tenfold increase, stating, ‘The last time I saw this speed was during the 3G era with mobile data.’ Zhang Peng explained the logic behind the recent price increase of Zhiyu GLM5 Turbo from the perspective of model manufacturers: Agents are no longer just question-and-answer; the token consumption for completing a task may be ten or even a hundred times that of answering simple questions, involving long-term task planning, continuous debugging, and multimodal information processing. ‘Long-term reliance on low-price competition is detrimental to the development of the entire industry.’ Luo Fuli believes the core bottleneck in the Agent era is the cost and speed of reasoning with long contexts, stating that only by achieving low costs and high speeds at the scale of millions or even tens of millions of tokens can truly high productivity tasks be assigned to models. She revealed that the model’s ‘self-evolution’ has moved from concept to practice, as models can now autonomously run for two to three days on clearly defined scientific research tasks, accelerating the team’s research efficiency by nearly ten times. She also posed a question: ‘The demand for reasoning has increased nearly tenfold in recent times; will the overall token growth this year reach a hundredfold?’ Huang Chao dissected three technical bottlenecks at the Agent application level: insufficient planning capabilities for long-chain tasks, memory inflation pressure caused by multi-Agent collaboration, and risks of low-quality tools and malicious injections in the skill ecosystem. He pointed out that the memory management of existing frameworks still relies on file systems and Markdown formats, and future developments need to move towards layered design. At the end of the roundtable, the four guests summarized the trends for the next 12 months with one word each: 1. Huang Chao: ‘Ecosystem,’ indicating that software will shift from human-oriented to Agent Native design. 2. Luo Fuli: ‘Self-evolution,’ calling it ‘the only place that can create new things.’ 3. Xia Lixue: ‘Sustainable tokens,’ hoping that China will become the world’s token factory. 4. Zhang Peng: ‘Computing power,’ stating that behind the tenfold growth, ‘there is still a hundredfold demand that has not been met.’