> ## Documentation Index
> Fetch the complete documentation index at: https://docs-d0.getdonut.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Trading Speed

> D0 moves instantly on routine work and switches to its strongest models for anything that needs real thinking. You never pick the mode yourself.

D0 runs two modes depending on what you're doing. For things you do constantly (checking prices, depositing, confirming trades) it moves instantly. For things that require real thinking (research, portfolio analysis, strategy backtesting) it switches to its strongest models and takes the time to do it properly.

You never choose which mode. D0 figures it out based on what you asked for.

## The fast lane

These are the actions you do dozens of times a day. D0 handles them without any AI reasoning step, so there is zero wait time.

### Slash commands: seven actions, one keystroke

Type **/** in the chat composer and the shortcuts appear:

| Command | What it does |
| - | - |
| **/trade** | Open a trade ticket |
| **/swap** | Swap one token for another |
| **/transfer** | Send funds to an address |
| **/balance** | Show wallet balances |
| **/positions** | View open positions |
| **/alert** | Set a price alert |
| **/research** | Research a token or market |

### Instant deposit address

Open **Portfolio** and click **Deposit**. The address and QR code for the venue appear immediately. D0 doesn't need to think about this one: deposit addresses are pre-computed.

### Direct execution

When the Trade modal or a chat quote is in front of you and you click the execute button, it goes straight to the venue. No AI step in between. The quote is already generated and the route is already selected; your click is the last human gate before it fires. **Execute LONG** or **Execute SHORT** on a signal card works the same way: the order modal opens prefilled, you confirm, it fires.

## Deep mode

When you need real analysis D0 switches to its strongest models automatically. You don't ask for it. It happens when the complexity of your request calls for it.

**What triggers deep mode:** research reports, multi-asset analysis, strategy backtesting, market regime breakdowns, cross-chain comparisons, anything that starts with "explain why."

The output is a full report: K-line charts, volume profiles, on-chain metrics, technical indicators, funding rates, data tables, and a structured thesis with confidence scoring. It takes longer, and it is supposed to.

> **You:** "Give me deep research on Bitcoin"
>
> **D0:** *\[returns the full report, 15 to 30 seconds]*

Reports are kept in **D0 Activity** in the left panel, so you can come back to them without regenerating.

## Why this matters

The same chat handles both. You don't switch interfaces, you don't pick a "fast model" or "smart model" yourself, and you don't wait when you don't need to. D0 routes each request based on what actually needs to happen.

Speed where it matters. Depth where it counts.

## Related

* [Three Modes of Chat](https://docs-d0.getdonut.ai/chat/three-modes.md)
* [Market Analysis](https://docs-d0.getdonut.ai/features/market-analysis.md)
* [The Trade Panel](https://docs-d0.getdonut.ai/trade/trade-panel.md)


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