> ## Documentation index
> Fetch the complete documentation index at: https://arize.com/llms.txt
> Ask it with GET https://arize.com/ask?q=<keywords or question>. Short keywords rank best.
> Use these to discover all available pages before exploring further.

# Observe

> Trace a model call and send it to Phoenix, in Python or TypeScript.

Trace an OpenAI call and send it to Arize Phoenix. Background: [What are traces](/docs/phoenix/tracing/concepts-tracing/what-are-traces).

You need a running Phoenix; see [Start Phoenix](/docs/phoenix/get-started). The example calls OpenAI, so set `OPENAI_API_KEY` too. Have an existing app? [Let your coding agent set it up](/docs/phoenix/get-started#have-an-existing-app).

## Instrument OpenAI

Register Phoenix as the trace destination and turn on auto-instrumentation. Phoenix instruments every installed [OpenInference](/docs/phoenix/integrations) package it finds, OpenAI here.

<Tabs>
  <Tab title="Python">
    ```bash
    pip install -U arize-phoenix-otel openai openinference-instrumentation-openai
    ```

    {/* snippet: examples/quickstarts/python/observe.py#register */}
    ```python
    from phoenix.otel import register

    register(
        project_name="tracing-quickstart",
        protocol="http/protobuf",
        auto_instrument=True,
    )
    ```
  </Tab>

  <Tab title="TypeScript">
    ```bash
    npm install openai @arizeai/phoenix-otel @arizeai/openinference-instrumentation-openai
    ```

    Put this in `instrumentation.ts` and import it before anything else:

    {/* snippet: examples/quickstarts/typescript/instrumentation.ts#register */}
    ```typescript
    import OpenAI from "openai";
    import { OpenAIInstrumentation } from "@arizeai/openinference-instrumentation-openai";
    import { register, registerInstrumentations } from "@arizeai/phoenix-otel";

    export const provider = register({ projectName: "tracing-quickstart" });

    const instrumentation = new OpenAIInstrumentation();
    instrumentation.manuallyInstrument(OpenAI);
    registerInstrumentations({ instrumentations: [instrumentation] });
    ```
  </Tab>
</Tabs>

## Make a call

Any OpenAI call now produces a trace.

<Tabs>
  <Tab title="Python">
    {/* snippet: examples/quickstarts/python/observe.py#call */}
    ```python
    from openai import OpenAI

    client = OpenAI()
    client.chat.completions.create(
        model="gpt-5-mini",
        messages=[{"role": "user", "content": "Why did my invoice change?"}],
    )
    ```
  </Tab>

  <Tab title="TypeScript">
    {/* snippet: examples/quickstarts/typescript/app.ts#call */}
    ```typescript
    import "./instrumentation";
    import OpenAI from "openai";
    import { provider } from "./instrumentation";

    const client = new OpenAI();
    await client.chat.completions.create({
      model: "gpt-5-mini",
      messages: [{ role: "user", content: "Why did my invoice change?" }],
    });
    await provider.shutdown();
    ```
  </Tab>
</Tabs>

## Look at the trace

Open the `tracing-quickstart` project in Phoenix and click the trace: the prompt, response, tokens, and latency are on the span. To trace your own code around the LLM call, wrap it with a [decorator](/docs/phoenix/tracing/how-to-tracing/setup-tracing/instrument#using-decorators) or [`traceChain`](/docs/phoenix/tracing/how-to-tracing/setup-tracing/setup-using-phoenix-otel).

## Learn more about traces

Other ways to read your traces:

- **In code**, with [Extract data from spans](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces/extract-data-from-spans).
- **From a coding agent**, with the [CLI](/docs/phoenix/integrations/developer-tools/cli) or [MCP](/docs/phoenix/integrations/mcp).
- **By asking [PXI](/docs/phoenix/pxi)**, for example *"why did this trace fail?"*
- **Across multi-turn conversations**, with [Sessions](/docs/phoenix/tracing/llm-traces/sessions).

To go deeper on tracing, the [Tracing Tutorial](/docs/phoenix/tracing/tutorial) covers how to read spans, understand timing, and use trace data to debug and analyze your agent. The code on this page lives in [examples/quickstarts](https://github.com/Arize-ai/phoenix/tree/main/examples/quickstarts), ready to run.