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

# BeeAI

> Trace BeeAI Framework agents in Python with OpenInference and send spans to Arize AX for LLM observability.

[BeeAI Framework](https://github.com/i-am-bee/beeai-framework) is a Python / TypeScript framework from IBM for building production-grade AI agents — tools, memory, multi-step reasoning, and pluggable LLM backends. Arize AX captures every BeeAI agent run via the [`openinference-instrumentation-beeai`](https://github.com/Arize-ai/openinference/tree/main/python/instrumentation/openinference-instrumentation-beeai) package.

## Prerequisites

* Python 3.11+
* An Arize AX account ([sign up](https://arize.com/sign-up/))
* An `OPENAI_API_KEY` from the [OpenAI Platform](https://platform.openai.com/api-keys)

## Launch Arize AX

1. Sign in to your [Arize AX account](https://app.arize.com/).
2. From **Space Settings**, copy your **Space ID** and **API Key**. You will set them as `ARIZE_SPACE_ID` and `ARIZE_API_KEY` below.

## Install

```bash theme={null}
pip install arize-otel openinference-instrumentation-beeai beeai-framework
```

## Configure credentials

```bash theme={null}
export ARIZE_SPACE_ID="<your-space-id>"
export ARIZE_API_KEY="<your-api-key>"
export ARIZE_PROJECT_NAME="beeai-tracing-example"
export OPENAI_API_KEY="<your-openai-api-key>"
```

## Setup tracing

```python theme={null}
# instrumentation.py
import os

from arize.otel import register
from openinference.instrumentation.beeai import BeeAIInstrumentor

tracer_provider = register(
    space_id=os.environ["ARIZE_SPACE_ID"],
    api_key=os.environ["ARIZE_API_KEY"],
    project_name=os.environ["ARIZE_PROJECT_NAME"],
)

BeeAIInstrumentor().instrument(tracer_provider=tracer_provider)
print("Arize AX tracing initialized for BeeAI.")
```

## Run BeeAI

```python theme={null}
# example.py

# Importing instrumentation first ensures tracing is set up
# before `beeai_framework` is imported.
from instrumentation import tracer_provider

import asyncio

from beeai_framework.agents.requirement.agent import RequirementAgent
from beeai_framework.backend.chat import ChatModel


async def main() -> None:
    # ChatModel.from_name routes through LiteLLM, so any
    # `<provider>:<model>` slug it supports works here. The OpenAI
    # provider reads OPENAI_API_KEY from the environment.
    llm = ChatModel.from_name("openai:gpt-5")

    agent = RequirementAgent(llm=llm, tools=[])
    response = await agent.run(
        "Why is the ocean salty? Answer in two sentences."
    )

    # response.output is a list of messages — print the final one.
    print(response.output[-1].text)


asyncio.run(main())
```

### Expected output

```text wrap theme={null}
Arize AX tracing initialized for BeeAI.
The ocean is salty because rivers continuously dissolve mineral salts from rocks and soil and carry them to the sea, where they accumulate over millions of years. Water leaves the ocean through evaporation but the salts remain, steadily concentrating until reaching today's roughly 3.5% salinity.
```

## Verify in Arize AX

1. Open your Arize AX space and select project **`beeai-tracing-example`**.
2. You should see a new trace within \~30 seconds with this shape: a `RequirementAgent` root span (AGENT) wraps an `OpenAIChatModel` LLM child span (model `gpt-5`, prompt + response + token usage attached) and a `final_answer` tool span.
3. If no traces appear, see [Troubleshooting](#troubleshooting).

## Troubleshooting

* **No traces in Arize AX.** Confirm `ARIZE_SPACE_ID` and `ARIZE_API_KEY` are set in the same shell that runs `example.py`. Enable OpenTelemetry debug logs with `export OTEL_LOG_LEVEL=debug` and re-run.
* **BeeAI spans missing but other spans present.** `BeeAIInstrumentor().instrument(...)` must run before any `from beeai_framework import ...`. Make sure `instrumentation.py` is the first import in your entry point.
* **`401` from OpenAI.** Verify `OPENAI_API_KEY` is set and has access to `gpt-5`. Swap the `openai:gpt-5` slug in `ChatModel.from_name(...)` for a model your key can call.
* **Other LLM providers.** BeeAI delegates model calls to LiteLLM, so any LiteLLM-supported provider works — `ChatModel.from_name("anthropic:claude-3-5-sonnet-20241022")`, `ChatModel.from_name("groq:llama-3.3-70b-versatile")`, `ChatModel.from_name("ollama:granite3.1-dense:8b")`, etc. The same `BeeAIInstrumentor` covers them.

## Resources

<CardGroup>
  <Card icon="book-open" href="https://framework.beeai.dev/" title="BeeAI Framework Documentation" horizontal />

  <Card icon="terminal" href="https://github.com/Arize-ai/openinference/tree/main/python/instrumentation/openinference-instrumentation-beeai" title="OpenInference BeeAI Instrumentor (Python)" horizontal />

  <Card icon="github" href="https://github.com/i-am-bee/beeai-framework" title="BeeAI Framework GitHub" horizontal />
</CardGroup>