Stanford 论文提出 Joules-to-GDP：衡量每焦耳算力产出的经济价值
Stanford 搞了个新指标算 AI 每焦耳能赚多少钱，结果小模型反而更划算，8 张 H100 只比 1 张快 22%，选模型前值得看看这套思路。
Stanford 研究人员构建了 Joules-to-GDP 指标，计算 AI 每消耗一焦耳能量所产出的净美元价值。他们在 GDPval 的 220 个专业任务（覆盖 44 个职业）上测试了 5 个开源模型。结果显示 GLM 5.2 单任务净价值最高，但 Gemma 4 31B 用 23% 的能量达到了它 56% 的价值。Qwen3.6-27B 从 1 张 H100 扩到 8 张，任务时间只缩短 22%，能耗却是 3.2 倍。GPT-OSS 120B 在价格和速度接近的情况下，能耗比 GLM 5.2 低 68%。
New Stanford paper finds that a smaller AI model can return far more economic value per unit of energy than the most capable one.
Now that the world is limited by energy/power, how much energy each AI task burns should indeed by studied.
Here, the researchers built Joules-to-GDP, which measures the dollar value of an AI's work, minus its running cost, per joule of energy used. They tested 5 open models on 220 professional tasks from GDPval, covering 44 occupations.
> GLM 5.2 produced the most net value per task, but Gemma 4 31B got 56% of it with 23% of the energy.
> Moving Qwen3.6-27B from 1 to 8 H100s cut task time by just 22% while using 3.2x the energy.
> GPT-OSS 120B used 68% less energy than GLM 5.2 despite nearly the same price and speed.
> Running the same model on 8 H100 chips instead of 1 made tasks only 22% faster but used 3.24x the energy, for nearly the same dollar value of work
https://t.co/KAj3vxhHto
