> ## Documentation Index
> Fetch the complete documentation index at: https://docs.orq.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Quick start guide
> Get started with Orq.ai in 5 minutes. Create your first deployment, make LLM API calls, and start building AI applications with step-by-step instructions.
## Welcome to [**orq.ai**](http://orq.ai)
This page covers the fastest way to create an initial deployment and make a first LLM call through the platform.
### Prerequisite
An account within the [**AI Studio**](https://my.orq.ai) is required.
[Sign Up Here](https://orq.ai/create-account) to create an orq.ai Account.
## Step 1: Setup an API Key with the available providers
To start using models, first add an API key with the chosen provider. To do so, open the [**AI Router > Providers**](/docs/integrations/providers/overview) tab, select a provider, and choose **Setup own API Key**.
To learn more about setting up Integrations, see [API keys & Endpoints](/docs/administer/api-keys).
## Step 2: Enable models in the AI Router
Browse the [**AI Router**](/docs/model-garden/overview) by going to the **AI Router** tab.
Here, find all models available to use within **orq.ai**, including their capabilities and a short description. To make any model available, **toggle the button** next to its name, it will then be usable in a first [Deployment](/docs/deployments/overview). To use **custom API keys**, head to the [Providers](/docs/router/providers-overview) section.
To learn more about all the models available, see [AI Router](/docs/model-garden/overview).
## Step 3: Set up a first Deployment
To create a first deployment, head to the **Deployment** tab, then select **Create Deployment**.
The Variant configuration screen will then appear. Here, select which parameters to use with the model and include the prompt(s) that will be sent to the language model.
Once ready, press the **Deploy** button at the top-right to save the latest configuration and make it available through the API.
There are many more features available to configure a Deployment. Including setting up multiple models, fallbacks, function calling, etc. To learn more, see [Creating a Deployment](/docs/deployments/creating).
## Step 4: Install the orq.ai SDK
Within the Deployment page, open the **code snippet** panel to find instructions to install and run the SDK for the current deployment.
Then, proceed to install the chosen SDK. Execute the following in a terminal to download the latest version of the SDK on the machine.
```bash Bash theme={"theme":{"light":"github-light","dark":"github-dark"}}
npm install @orq-ai/node --save
```
```bash Bash theme={"theme":{"light":"github-light","dark":"github-dark"}}
!pip install orq-ai-sdk
```
## Step 5: Get an API Token
An API key is needed to set up the client. This can be found within the **Workspace Settings > API Keys** menu.
To learn more about API Authentication, see [Authentication](/docs/administer/api-keys).
## Step 6: Use the SDK to invoke a Deployment
Once the API Token has been saved for use, refer back to the code snippet accessed before the Deployment to fetch the code to initialize and invoke a deployment:
> Don't forget to replace `` with the API key fetched previously.
```typescript TypeScript theme={"theme":{"light":"github-light","dark":"github-dark"}}
import { Orq } from "@orq-ai/node";
const client = new Orq({
apiKey: '',
environment: 'production',
});
const deployment = await client.deployments.invoke({
key: "quickstart-deployment",
metadata: {
"custom-field-name": "custom-metadata-value"
}
});
```
```python Python theme={"theme":{"light":"github-light","dark":"github-dark"}}
import os
from orq_ai_sdk import Orq
client = Orq(
api_key=os.environ.get("ORQ_API_KEY", ""),
environment="production"
)
generation = client.deployments.invoke(
key="quickstart-deployment",
metadata={
"custom-field-name": "custom-metadata-value"
}
)
print(generation.choices[0].message.content)
```
Congratulations, the first deployment invocation with **orq.ai** was successfully run. To learn more about deployments and all the possibilities they offer, see [Creating a Deployment](/docs/deployments/creating).
## Step 7: Track projects and costs
**orq.ai** allows tracking costs and projects grouped in [Identities](/docs/analytics/identity).
Pass an `identity_id` / `identityId` when initializing the client to associate all usage with a specific identity.
> Identities can also be created using the SDK. See how with the [Node SDK](https://github.com/orq-ai/orq-node/blob/main/docs/sdks/identities/README.md#create) and [Python SDK](https://github.com/orq-ai/orq-python/blob/main/docs/sdks/identities/README.md#create).
```typescript TypeScript theme={"theme":{"light":"github-light","dark":"github-dark"}}
import { Orq } from "@orq-ai/node";
const client = new Orq({
apiKey: '',
environment: 'production',
identityId: ''
});
const deployment = await client.deployments.invoke({
key: "quickstart-deployment",
metadata: {
"custom-field-name": "custom-metadata-value"
}
});
```
```python Python theme={"theme":{"light":"github-light","dark":"github-dark"}}
import os
from orq_ai_sdk import Orq
client = Orq(
api_key=os.environ.get("ORQ_API_KEY", ""),
environment="production",
identity_id=""
)
generation = client.deployments.invoke(
key="quickstart-deployment",
metadata={
"custom-field-name": "custom-metadata-value"
}
)
print(generation.choices[0].message.content)
```
To learn more about Identities and the resulting Analytics, see [Using Identity Metrics](/docs/analytics/identity-metrics).
The Variant configuration screen will then appear. Here, select which parameters to use with the model and include the prompt(s) that will be sent to the language model.
Once ready, press the **Deploy** button at the top-right to save the latest configuration and make it available through the API.
Then, proceed to install the chosen SDK. Execute the following in a terminal to download the latest version of the SDK on the machine.
**orq.ai** allows tracking costs and projects grouped in [Identities](/docs/analytics/identity).
Pass an `identity_id` / `identityId` when initializing the client to associate all usage with a specific identity.
> Identities can also be created using the SDK. See how with the [Node SDK](https://github.com/orq-ai/orq-node/blob/main/docs/sdks/identities/README.md#create) and [Python SDK](https://github.com/orq-ai/orq-python/blob/main/docs/sdks/identities/README.md#create).