30 August 2026

Google and Purdue reduce AI agent memory use by 94 percent

First reported

The Neuron ran this on 30 August.

Researchers created SKILL.state, a method that stores an AI agent's current status as structured data instead of replaying its entire conversation history.
On a 100-step benchmark test using Google's Gemini model, the technique cut token consumption (computational resources needed) by approximately 94 percent.
The approach also improved accuracy, meaning the agent made better decisions while using far fewer resources.

How it was covered

The NeuronPete Huang & Grant Harvey

Researchers at Google and Purdue introduced SKILL.state, which keeps an agent's current structured state instead of replaying full history, cutting token use about 94 percent on a 100-step Gemini benchmark while improving accuracy.