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

# Send to Grafana Cloud

> Configure the OpenLIT SDK to send AI observability data to Grafana Cloud using the OTLP gateway

Grafana Cloud accepts OpenTelemetry data through a managed OTLP gateway. You authenticate with a Base64-encoded Instance ID and API token passed as an `Authorization` header.

## Get started

<Steps>
  <Step title="Get your Grafana Cloud OTLP credentials">
    1. Sign in to the [Grafana Cloud Portal](https://grafana.com/auth/sign-in) and select your Grafana Cloud Stack.
    2. Click **Configure** in the OpenTelemetry section.
    3. In the **Password / API Token** section, click **Generate now** to create a new API token:
       * Give the token a name, for example `openlit`.
       * Click **Create token**, then **Close**.
    4. Copy and save the values for `OTEL_EXPORTER_OTLP_ENDPOINT` and `OTEL_EXPORTER_OTLP_HEADERS`.

    <Note>
      Replace the space after `Basic` in the headers value with `%20`:
      `OTEL_EXPORTER_OTLP_HEADERS="Authorization=Basic%20[base64 instanceID:token]"`
    </Note>
  </Step>

  <Step title="Install the OpenLIT SDK">
    <Tabs>
      <Tab title="Python">
        ```shell theme={null}
        pip install openlit
        ```
      </Tab>

      <Tab title="TypeScript">
        ```shell theme={null}
        npm install openlit
        ```
      </Tab>
    </Tabs>
  </Step>

  <Step title="Configure the SDK to send to Grafana Cloud">
    <Tabs>
      <Tab title="Python">
        **Using function arguments:**

        ```python theme={null}
        import openlit

        openlit.init(
            otlp_endpoint="https://otlp-gateway-<zone>.grafana.net/otlp",
            otlp_headers="Authorization=Basic%20<base64 instanceID:token>"
        )
        ```

        **Using environment variables:**

        ```shell theme={null}
        export OTEL_EXPORTER_OTLP_ENDPOINT="https://otlp-gateway-<zone>.grafana.net/otlp"
        export OTEL_EXPORTER_OTLP_HEADERS="Authorization=Basic%20<base64 instanceID:token>"
        ```

        ```python theme={null}
        import openlit

        openlit.init()
        ```
      </Tab>

      <Tab title="TypeScript">
        **Using function arguments:**

        ```typescript theme={null}
        import Openlit from 'openlit';

        Openlit.init({
            otlpEndpoint: 'https://otlp-gateway-<zone>.grafana.net/otlp',
            otlpHeaders: 'Authorization=Basic%20<base64 instanceID:token>'
        });
        ```

        **Using environment variables:**

        ```shell theme={null}
        export OTEL_EXPORTER_OTLP_ENDPOINT="https://otlp-gateway-<zone>.grafana.net/otlp"
        export OTEL_EXPORTER_OTLP_HEADERS="Authorization=Basic%20<base64 instanceID:token>"
        ```

        ```typescript theme={null}
        import Openlit from 'openlit';

        Openlit.init();
        ```
      </Tab>
    </Tabs>

    Replace:

    * `<zone>` with your Grafana Cloud stack zone (e.g., `prod-us-east-0`).
    * `<base64 instanceID:token>` with the Base64-encoded value of your Instance ID and API token.
  </Step>
</Steps>

## Configuration values

| Value                         | Where to find it                             |
| ----------------------------- | -------------------------------------------- |
| `OTEL_EXPORTER_OTLP_ENDPOINT` | Grafana Cloud Portal → OpenTelemetry section |
| `OTEL_EXPORTER_OTLP_HEADERS`  | Grafana Cloud Portal → OpenTelemetry section |

<Warning>
  The `Authorization` header value must URL-encode the space after `Basic` as `%20`. Using a literal space causes authentication failures.
</Warning>

## Full example

```python theme={null}
import openlit
import openai

openlit.init(
    otlp_endpoint="https://otlp-gateway-prod-us-east-0.grafana.net/otlp",
    otlp_headers="Authorization=Basic%20dXNlcjE6cGFzc3dvcmQ=",
    service_name="my-ai-service",
    environment="production"
)

client = openai.OpenAI()
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
```

***

<CardGroup cols={3}>
  <Card title="Send to OpenLIT" href="/sdk/destinations/openlit" icon="server">
    Send to a self-hosted OpenLIT platform instance
  </Card>

  <Card title="Send to Datadog" href="/sdk/destinations/datadog" icon="dog">
    Route traces through the Datadog Agent
  </Card>

  <Card title="Integrations" href="/sdk/integrations/overview" icon="circle-nodes">
    60+ LLM providers, AI frameworks, and vector databases
  </Card>
</CardGroup>