跳到正文
Rohan Paul@rohanpaul_ai
60AI 编辑部评分,满分 100

Stanford 与 Together AI 提出 Intelligence per Watt 指标度量本地 AI 能效

2026-09-11 14:43· 5小时前
AI 导读

Stanford 大学与 Together AI 发布论文《Intelligence per Watt: Measuring Intelligence Efficiency of Local AI》,提出以任务准确率除以功耗的 IPW 指标评估本地推理。

Absolutely beautiful new paper from Stanford Univ + Together AI

"Intelligence per Watt: Measuring Intelligence Efficiency of Local AI"

• It found that hybrid local-cloud routing reduced energy, compute, and cost by 60% to 80% against its batched-cloud baseline.

• From 2023 to 2025, local AI’s intelligence-per-watt improved 5.3×, while locally serviceable query coverage jumped from 23.2% to 71.3%.

• An iPhone 16 Pro achieved roughly 7× higher intelligence-per-watt than workstation GPUs on the same model and precision, showing how power-efficient mobile AI can be for lightweight queries.

• A diverse pool of 20+ local models actually beat the paper’s 3 frontier cloud models on 3 of 4 benchmarks when each query was routed to the best model, showing how much model diversity can matter.

• Dropping precision from FP16 to FP4 cut inference energy by 3X–3.5X, while costing roughly 2.5 percentage points of accuracy per precision step; in one test, a larger FP4 model even beat a smaller FP16 model.

• The biggest remaining weakness is concentrated at the hard end: on the paper’s hardest reasoning slice, about 95% of problems were still unsolved by local models, even as easier and medium-difficulty tasks improved rapidly.

来源:Rohan Paul· x.com