Google DeepMind’s new routing idea is trying to solve a great practical question.

Routing is supposed to save compute, but deciding where to route can itself cost compute.

Paper says spend more on model selection only when the better decision is worth the extra cost.

Google DeepMind frames that decision as a Pandora’s Box problem.

Every specialist gets a cheap, noisy score first. The router pays for a stronger estimate only when the expected value of that extra information is higher than its cost.

That changes the economics of routing.

Across MATH, RAG, and EmbedLLM, Pandora’s Router had the lowest or tied-lowest combined routing regret and inspection cost on average across the tested cost settings.

On EmbedLLM, with more than 100 routing targets, its average inspection cost was 0.075 versus 1.986 for always running the expensive estimator, while routing regret also fell from 0.370 to 0.311.

– arxiv. org/abs/2608.20316

Title: "Pandora's AI Model Routing Box: Efficient Allocation with Costly Value Estimation"