> ## Documentation Index
> Fetch the complete documentation index at: https://arizeai-433a7140.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# OrcaRouter Tracing

> Phoenix provides auto-instrumentation for OrcaRouter through the OpenAI Python Library since OrcaRouter provides a fully OpenAI-compatible API endpoint.

## Install

```bash theme={null}
pip install openinference-instrumentation-openai openai
```

## Setup

Add your OrcaRouter API key as an environment variable:

```bash theme={null}
export ORCAROUTER_API_KEY='your_orcarouter_api_key'
```

Use the register function to connect your application to Phoenix:

```python theme={null}
from phoenix.otel import register

# configure the Phoenix tracer
tracer_provider = register(
  project_name="my-llm-app", # Default is 'default'
  auto_instrument=True # Auto-instrument your app based on installed dependencies
)
```

## Run OrcaRouter

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

client = openai.OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key=os.environ["ORCAROUTER_API_KEY"]
)

response = client.chat.completions.create(
    model="orcarouter/auto",
    messages=[{"role": "user", "content": "Write a haiku about observability."}],
)
print(response.choices[0].message.content)
```

`orcarouter/auto` selects an upstream provider per request. You can also target a specific provider model using `<provider>/<model>` format (e.g. `openai/gpt-4.1-mini`, `anthropic/claude-haiku-4-5`). To test without a funded account, use a free model such as `deepseek/deepseek-v4-flash-free`.

## Observe

Now that you have tracing set up, all invocations of the OpenAI client pointed at OrcaRouter will be streamed to your running Phoenix for observability and evaluation.

## What Gets Traced

All OrcaRouter model calls are automatically traced and include:

* Request/response data and timing
* Model name — the resolved upstream model name (e.g. `deepseek-v4-flash-202505`), not the virtual `orcarouter/auto` identifier
* Token usage and cost data
* Error handling and debugging information

## Common Issues

* **API Key**: Use your OrcaRouter API key (`sk-orca-...`), not an OpenAI key
* **Model Names**: Use `orcarouter/auto` for adaptive routing, or `<provider>/<model>` for a specific upstream. See [OrcaRouter's documentation](https://docs.orcarouter.ai/introduction) for available models
* **Insufficient balance**: `orcarouter/auto` routes to paid upstream models and requires a funded OrcaRouter wallet. Use `deepseek/deepseek-v4-flash-free` to test without balance
* **Base URL**: Ensure you're using `https://api.orcarouter.ai/v1` as the base URL

## Resources

<Columns cols={2}>
  <Card title="OrcaRouter Documentation" href="https://docs.orcarouter.ai/introduction" icon="book" horizontal description="OrcaRouter setup docs" />

  <Card title="OpenInference OpenAI Instrumentation" href="https://github.com/Arize-ai/openinference/tree/main/python/instrumentation/openinference-instrumentation-openai" icon="puzzle-piece" horizontal description="OpenAI instrumentation package" />
</Columns>