AI市场重心在中间层：定价战与开源套利
The Most Important Market in AI is the Middle
用具体数据拆解了AI市场的真实需求结构（正态分布vs金字塔），揭示了中间层定价战的底层逻辑，对SaaS产品定位和成本控制有直接参考价值。
Yesterday, Anthropic released a new model & cut its price. Ninety minutes later, OpenAI did the same.
昨天，Anthropic 发布了一款新模型并降低了价格。90分钟后，OpenAI 也采取了同样的行动。
Most business AI use is the messy middle: multi-step workflows that need a smart enough model at a price a company can afford. It is the most important part of the market today, & it is where the competition is fiercest. The price cuts are the evidence.
大多数企业的 AI 应用处于‘混乱的中层’：需要足够智能的模型来处理多步骤工作流，且价格在公司可承受范围内。这是当今市场最重要的部分，也是竞争最激烈的领域。价格的下调就是证据。
In June, Anthropic set the frontier price at $10 & $50 per million tokens with Fable 5. In July, OpenAI answered with GPT-5.6 Sol at $5 & $30, matching that capability at a third of the cost per task.
6月，Anthropic 为 Fable 5 设定了前沿价格，每百万 token 分别为 10 美元和 50 美元。7月，OpenAI 以 GPT-5.6 Sol 回应，定价为每百万 token 5 美元和 30 美元，以三分之一的任务成本实现了相当的能力。
The Opus line had never moved. Opus 4.5, 4, 4.8 & 5 all listed at $5 & $25 per million tokens. Yesterday’s cut was the first.
Opus 系列的价格从未变动过。Opus 4.5、4、4.8 和 5 的标价均为每百万 token 5 美元和 25 美元。昨天的降价是首次调整。
It is even more extreme at the low end. OpenAI cut Luna by 80% in July, then cut it another 50% yesterday.
在低端市场的情况更为极端。OpenAI 在7月将 Luna 的价格降低了 80%，随后在昨天又将其降低了 50%。
More than just closed source rivalry, open models deflate prices too. The generics on the AI grocery aisle run a majority of token volume on the gateways that publish data, at an 86% discount to the blended price of closed models.
不仅仅是闭源模型之间的竞争，开源模型也在压低价格。在发布数据的网关上，AI 杂货货架上的通用模型占据了大部分 token 用量，其价格比闭源模型的混合价格低 86%。
Large customers pursue even greater savings with fine tuning. Cursor’s Composer 2 fine tuned Kimi K2.5, an open-weight base, cutting its overall cost 86% against its previous in-house model. Harvey did the same, cutting cost per cell 55% against Sonnet 5 while scoring higher than Fable 5.
大型客户通过微调追求更大的节省。Cursor 的 Composer 2 对 Kimi K2.5（一个开源权重基础模型）进行了微调，使其总成本较之前的内部模型降低了 86%。Harvey 也采取了类似做法，与 Sonnet 5 相比，其每个单元的成本降低了 55%，同时得分高于 Fable 5。
But the right tail of the market is thinner than almost anyone forecast. Anthropic’s Fable 5.1, its most capable & most expensive model, commanded only 3.7% of gateway spending in its first twelve days. Its predecessor peaked at 13.2% when access was restored in July, then fell to 4.9% a month later when Opus 5 shipped at half the price. Among large corporate accounts, frontier models fell from 53% of token consumption in early August to 45% by September.
但市场右侧的尾部比几乎所有人的预测都要薄。Anthropic 最具能力且最昂贵的模型 Fable 5.1，在其上线的前十二天内仅占网关支出的 3.7%。其前身在今年7月恢复访问时峰值达到 13.2%，随后在一个月后 Opus 5 以半价推出时降至 4.9%。在大型企业账户中，前沿模型在8月初占 token 消费量的 53%，到9月已降至 45%。
Demand for intelligence is not a pyramid with a small, wealthy peak paying for everything beneath it. It is a normal distribution with a fat middle. The middle buys intelligence per dollar.
对智能的需求并非一座金字塔，即由顶部少量富裕用户支付底部所有费用。它呈正态分布，中间部分较为厚实。中层用户按每美元购买智能。
Intelligence costs keep plummeting. What enterprises demand from AI does not change nearly as fast. So the tier that satisfies a fixed requirement keeps getting cheaper.
智能的成本持续暴跌。企业对 AI 的需求变化远没有这么快。因此，满足固定需求的层级变得越来越便宜。
As intelligence per dollar explodes, the distribution of tokens may shift to commodity. Whether that happens will determine the economics of the AI market.
随着每美元获得的智能量激增，token 的分布可能会转向大宗商品化。这是否会发生将决定 AI 市场的经济格局。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力