So we started to think about what work went where.
Drew 的结论落在任务分流：按难度把活派给不同价位的模型，本质上是把云计算的实例选型搬到了模型层。但前提是你得能提前判断一个任务有多难，而这恰恰是当前最缺的评估能力——分错档的代价会同时体现在成本和返工上。
3 Matching Annotations
- Last 7 days
-
simonwillison.net simonwillison.net
-
-
A new model would arrive at the same price (or cheaper!) and paper over most of your problems.
Drew Breunig 点破了过去三年的隐性补贴：工程上的糙活由下一代模型免费兜底，所以打磨 harness 和上下文策略显得像过度优化。这个循环一断，精细化工程第一次有了明确回报期——它从可选项变回了竞争壁垒。
-
-
simonwillison.net simonwillison.net
-
which looks reasonable given that Opus 5 was only released on July 24th, and supports the idea that Fable's cost has made it a less popular model
顶配模型份额垫底、上一代 Opus 4.8 独占近三成，这恰好是 Drew Breunig 说的"免费午餐结束"的市场侧证据：最强模型贵到不能全量使用，工程团队就被迫自己做任务分流。性价比模型胜出不代表能力过剩，而是使用方式变了。
-