> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ateve.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Token efficiency

> Query-aware context compression slashes token usage without sacrificing quality.

## Benchmark results

<div className="ateve-token-benchmark">
  <img src="https://mintcdn.com/ateve-f09deb12/k0HIIkRnF5Fo5r8z/images/benchmarks/token-efficiency.svg?fit=max&auto=format&n=k0HIIkRnF5Fo5r8z&q=85&s=015ea4358eef44a1eb6cda6ca2931d2f" alt="Ateve token efficiency comparison from the official website" width="693" height="557" data-path="images/benchmarks/token-efficiency.svg" />
</div>

## About this methodology

Ateve dynamically combines Chunk, Fact, and Entity indexes based on query intent, retrieving only high-relevance, answer-ready context rather than full raw pages. This minimizes token overhead while preserving retrieval quality.

## Read more

<Card title="Ateve Blog" href="https://ateve.ai/blog">
  Explore Ateve's benchmark articles and evaluation methodologies.
</Card>