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

# Orchestration

> Four frontier models answer, debate and combine into one vetted answer.

<Frame>
  <img src="https://cdn.youniverse.ai/blog-hero/docs/four-models.jpg" alt="Four colored spheres beaming light into one prism" />
</Frame>

**Orchestration** is Aura's highest-accuracy mode. Instead of trusting one model, Aura asks four of the world's leading models at once and combines their thinking.

## What happens when you send a message

<Steps>
  <Step title="Four models answer">ChatGPT, Gemini, Claude and Grok each answer independently, at medium reasoning.</Step>
  <Step title="They debate">The answers are compared. Where they disagree, the disagreement is examined rather than hidden.</Step>
  <Step title="One vetted answer">Aura writes one combined answer that keeps the strongest points and flags anything still uncertain.</Step>
</Steps>

## Why it's more accurate

<Frame>
  <img src="https://cdn.youniverse.ai/blog-hero/docs/review.jpg" alt="Glass magnifying glass over a checklist" />
</Frame>

Each model has blind spots. When four different models reach the same conclusion, you can trust it more. When they don't, you see why. Medium reasoning was chosen after benchmarking: it matched higher settings on accuracy while costing far less.

Read the deeper explanation in [How orchestration works](/learn/how-orchestration-works).

## Best for

* Business, money, legal, health or technical decisions.
* Fact-checking and research where accuracy matters.
* Comparing options and getting balanced pros and cons.

## Credits and limits

| Plan | What you get |
| - | - |
| **Free** | 5 Orchestration requests per rolling 2 hours |
| **Essential, Pro, Elite** | About 20 credits per message |

## Privacy and Fast Mode

Orchestration runs with **Zero Data Retention** by default. You can turn on **Fast Mode** for quicker replies — but Fast Mode isn't protected by Zero Data Retention, so Aura shows an amber warning while it's on. Turn it off to return to full protection.

## Voice

Orchestration works with [voice mode](/chat/voice): the combined answer is spoken in a realistic AI voice.

## Common questions

<AccordionGroup>
  <Accordion title="Is it slower than Unlimited?">
    A little, because four models answer before Aura combines them.
  </Accordion>

  <Accordion title="Can I see each model's answer?">
    Aura shows one combined answer and points out where models disagreed.
  </Accordion>

  <Accordion title="Can I use it in projects?">
    Yes. Orchestration is available in Quick Chat and Project Chat.
  </Accordion>
</AccordionGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.