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

# Cloudflare

> Trace AI requests through Cloudflare AI Gateway and export Cloudflare Agents traces to Braintrust with OpenTelemetry

If you are a coding agent, prefer the Braintrust [`bt` CLI](/docs/reference/cli/quickstart) for repeatable, scriptable work: running evals, instrumenting code, querying logs, syncing data, managing functions, and configuring coding agents. Use the MCP server for reasoning over Braintrust data in conversation, such as ad-hoc lookups and exploration from your IDE.

Braintrust integrates with Cloudflare in two ways: tracing LLM requests routed through [Cloudflare AI Gateway](https://developers.cloudflare.com/ai-gateway/), and exporting traces from [Cloudflare Agents](https://developers.cloudflare.com/agents/) over OpenTelemetry. This page covers both.

## Cloudflare AI Gateway

[Cloudflare AI Gateway](https://developers.cloudflare.com/ai-gateway/) provides a unified interface to access multiple AI providers with observability, caching, and rate limiting. Braintrust automatically traces requests through Cloudflare AI Gateway across all supported providers.

### Setup

Install the Braintrust SDK and OpenAI client:

<CodeGroup>
  ```bash Typescript theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
  # pnpm
  pnpm add braintrust openai
  # npm
  npm install braintrust openai
  ```

  ```bash Python theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
  pip install braintrust openai
  ```
</CodeGroup>

Configure your environment variables:

```bash title=".env" theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
OPENAI_API_KEY=<your-provider-api-key>
CLOUDFLARE_ACCOUNT_ID=<your-cloudflare-account-id>
CLOUDFLARE_AI_GATEWAY_NAME=<your-gateway-name>
BRAINTRUST_API_KEY=<your-braintrust-api-key>
```

### Trace with Cloudflare AI Gateway

Use `wrapOpenAI` to automatically trace requests through Cloudflare's unified API endpoint:

<CodeGroup dropdown>
  ```typescript theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
  import { OpenAI } from "openai";
  import { initLogger, wrapOpenAI } from "braintrust";

  // Initialize Braintrust logging
  initLogger({
    projectName: "My Project",
    apiKey: process.env.BRAINTRUST_API_KEY,
  });

  // Create OpenAI client configured for Cloudflare AI Gateway
  const client = wrapOpenAI(
    new OpenAI({
      // OpenAI SDK automatically adds /chat/completions
      baseURL: `https://gateway.ai.cloudflare.com/v1/${process.env.CLOUDFLARE_ACCOUNT_ID}/${process.env.CLOUDFLARE_AI_GATEWAY_NAME}/compat`,
      apiKey: process.env.OPENAI_API_KEY,
    }),
  );

  // This request will be automatically traced by Braintrust
  const result = await client.chat.completions.create({
    model: "openai/gpt-4o",
    messages: [{ role: "user", content: "What is 1+1?" }],
  });

  console.log(result);
  ```

  ```python theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
  import os
  import openai
  from braintrust import init_logger, wrap_openai

  # Initialize Braintrust logging
  logger = init_logger(project="My Project")

  # Create OpenAI client configured for Cloudflare AI Gateway
  client = wrap_openai(
      # OpenAI client automatically adds /chat/completions
      openai.OpenAI(
          base_url=f"https://gateway.ai.cloudflare.com/v1/{os.getenv('CLOUDFLARE_ACCOUNT_ID')}/{os.getenv('CLOUDFLARE_AI_GATEWAY_NAME')}/compat",
          api_key=os.getenv("OPENAI_API_KEY"),
      )
  )

  # This request will be automatically traced by Braintrust
  result = client.chat.completions.create(
      model="openai/gpt-4o",
      messages=[{"role": "user", "content": "What is 1+1?"}],
  )

  print(result)
  ```
</CodeGroup>

### Switching providers

Cloudflare AI Gateway's unified API allows you to easily switch between different AI providers by changing the `model` parameter and corresponding API key. All requests are automatically traced by Braintrust regardless of which provider you use.

<CodeGroup dropdown>
  ```typescript theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
  import { OpenAI } from "openai";
  import { wrapOpenAI } from "braintrust";

  // Switch to Anthropic
  const client = wrapOpenAI(
    new OpenAI({
      baseURL: `https://gateway.ai.cloudflare.com/v1/${process.env.CLOUDFLARE_ACCOUNT_ID}/${process.env.CLOUDFLARE_AI_GATEWAY_NAME}/compat`,
      apiKey: process.env.ANTHROPIC_API_KEY, // Use Anthropic's API key
    }),
  );

  const result = await client.chat.completions.create({
    model: "anthropic/claude-sonnet-4-5-20250929", // Use Anthropic model
    messages: [{ role: "user", content: "Hello!" }],
  });
  ```

  ```python theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
  import os
  import openai
  from braintrust import wrap_openai

  # Switch to Anthropic
  client = wrap_openai(
      openai.OpenAI(
          base_url=f"https://gateway.ai.cloudflare.com/v1/{os.getenv('CLOUDFLARE_ACCOUNT_ID')}/{os.getenv('CLOUDFLARE_AI_GATEWAY_NAME')}/compat",
          api_key=os.getenv("ANTHROPIC_API_KEY"),  # Use Anthropic's API key
      )
  )

  result = client.chat.completions.create(
      model="anthropic/claude-sonnet-4-5-20250929",  # Use Anthropic model
      messages=[{"role": "user", "content": "Hello!"}],
  )
  ```
</CodeGroup>

Cloudflare AI Gateway supports OpenAI, Anthropic, Google AI Studio, Groq, Mistral, Cohere, Perplexity, DeepSeek, Cerebras, xAI, and Workers AI. See the [Cloudflare documentation](https://developers.cloudflare.com/ai-gateway/usage/chat-completion/) for the complete list.

### AI Gateway resources

* [Cloudflare AI Gateway documentation](https://developers.cloudflare.com/ai-gateway/)
* [Supported providers](https://developers.cloudflare.com/ai-gateway/usage/chat-completion/)

## Cloudflare Agents

[Cloudflare Agents](https://developers.cloudflare.com/agents/) instrument each agent run with OpenTelemetry spans that follow the [GenAI semantic conventions](https://github.com/open-telemetry/semantic-conventions-genai/blob/main/docs/gen-ai/gen-ai-spans.md). Because Braintrust ingests OpenTelemetry data, you can export these spans to Braintrust from Cloudflare's [Workers Observability](https://developers.cloudflare.com/workers/observability/exporting-opentelemetry-data/) without adding the Braintrust SDK to your Worker. Braintrust maps the incoming spans to `LLM` and tool spans automatically.

<Steps>
  <Step title="Get your Braintrust credentials">
    Create an API key in **<Icon icon="settings-2" /> Settings > [<Icon icon="key-square" /> API keys](https://www.braintrust.dev/app/~/configuration/org/api-keys)**. Find your project ID using the **Copy project ID** button at the bottom of the project's configuration page.
  </Step>

  <Step title="Add a tracing destination in Cloudflare">
    In the Cloudflare dashboard, go to your account's **Workers Observability** section and add a destination pointing at Braintrust's OpenTelemetry endpoint. Name it `braintrust` so you can reference it in the next step:

    * **Name**: `braintrust`.
    * **OTLP endpoint**: `https://api.braintrust.dev/otel/v1/traces`.
    * **Custom headers**:
      * `Authorization`: `Bearer <BRAINTRUST_API_KEY>`.
      * `x-bt-parent`: `project_id:<BRAINTRUST_PROJECT_ID>`.

    The `x-bt-parent` header sets the project or experiment that receives the traces. For other parent formats and the full attribute mapping, see the [OpenTelemetry integration](/docs/integrations/sdk-integrations/opentelemetry#otlp-configuration).

    <Note>
      If your organization is on the EU data plane, use `https://api-eu.braintrust.dev/otel/v1/traces` instead. Self-hosted deployments should use their stack's API URL followed by `/otel/v1/traces`. See [Data plane region](/docs/admin/organizations#data-plane-region).
    </Note>
  </Step>

  <Step title="Enable trace export in wrangler">
    In your Worker's `wrangler.jsonc`, enable trace export and reference the `braintrust` destination you created by name:

    ```json title="wrangler.jsonc" theme={"theme":{"light":"github-light","dark":"github-dark-dimmed"}}
    {
      "observability": {
        "traces": {
          "enabled": true,
          "destinations": ["braintrust"]
        }
      }
    }
    ```

    Deploy your Worker. Agent runs now appear as traces in Braintrust.
  </Step>
</Steps>

### Cloudflare Agents resources

* [Cloudflare Agents tracing](https://developers.cloudflare.com/agents/runtime/operations/observability/tracing/#exporting-traces)
* [Exporting OpenTelemetry data from Workers](https://developers.cloudflare.com/workers/observability/exporting-opentelemetry-data/)
* [Braintrust OpenTelemetry integration](/docs/integrations/sdk-integrations/opentelemetry)