# How Derived Views in Connect AI Reduce LLM Token Usage

**URL:** <https://community.cdata.com/t/how-derived-views-in-connect-ai-reduce-llm-token-usage/1256>\
**Category:** CData Connect AI\
**Tags:** ai-integration, cdata-connect-ai, llm-optimization, ai-analytics, derived-views, token-reduction, multi-source, llm-token, discussion, ai\
**Created:** [December 23, 2025, 7:24pm UTC](https://community.cdata.com/t/how-derived-views-in-connect-ai-reduce-llm-token-usage/1256 "2025-12-23T19:24:19Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Cameron\_L](https://avatars.discourse-cdn.com/v4/letter/c/e95f7d/32.png) [@Cameron\_L](https://community.cdata.com/u/Cameron_L)\
**Post date:** [December 23, 2025, 7:24pm UTC](https://community.cdata.com/t/how-derived-views-in-connect-ai-reduce-llm-token-usage/1256/1 "2025-12-23T19:24:20Z")

</div>

AI-driven analytics can be slow and costly when LLMs repeatedly rebuild schemas and relationships across multiple data sources. Derived Views with CData Connect AI eliminate that overhead, reducing token usage while making multi-source analytics faster and more reliable at scale.

Read the full article here:&nbsp;[https://www.cdata.com/kb/articles/connect-ai-derived-views-performance.rst](https://www.cdata.com/kb/articles/connect-ai-derived-views-performance.rst)