> ## 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.
# Smolagents
> Trace Hugging Face smolagents with OpenInference and send spans to Arize AX for LLM observability.
[smolagents](https://github.com/huggingface/smolagents) is Hugging Face's minimalist Python agent framework — a few hundred lines of code-first agents with tools, memory, and multi-step reasoning. Arize AX captures every smolagent run — agent steps, tool invocations, and the underlying LLM calls — via the [`openinference-instrumentation-smolagents`](https://github.com/Arize-ai/openinference/tree/main/python/instrumentation/openinference-instrumentation-smolagents) package.
## Prerequisites
* Python 3.10+
* 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-smolagents \
smolagents openai
```
## Configure credentials
```bash theme={null}
export ARIZE_SPACE_ID=""
export ARIZE_API_KEY=""
export ARIZE_PROJECT_NAME="smolagents-tracing-example"
export OPENAI_API_KEY=""
```
## Setup tracing
```python theme={null}
# instrumentation.py
import os
from arize.otel import register
from openinference.instrumentation.smolagents import SmolagentsInstrumentor
tracer_provider = register(
space_id=os.environ["ARIZE_SPACE_ID"],
api_key=os.environ["ARIZE_API_KEY"],
project_name=os.environ["ARIZE_PROJECT_NAME"],
)
SmolagentsInstrumentor().instrument(tracer_provider=tracer_provider)
print("Arize AX tracing initialized for smolagents.")
```
## Run smolagents
```python theme={null}
# example.py
# Importing instrumentation first ensures tracing is set up
# before `smolagents` is imported.
from instrumentation import tracer_provider
from smolagents import CodeAgent, OpenAIServerModel
# OpenAIServerModel reads OPENAI_API_KEY from the environment.
model = OpenAIServerModel(model_id="gpt-5")
agent = CodeAgent(tools=[], model=model)
result = agent.run("Why is the ocean salty? Answer in two sentences.")
print(result)
```
### Expected output
```text wrap theme={null}
Arize AX tracing initialized for smolagents.
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 **`smolagents-tracing-example`**.
2. You should see a new trace within \~30 seconds containing a `CodeAgent.run` parent span wrapping per-step `Step N` spans, an `OpenAIModel.generate` LLM child span (with the prompt, response, and token usage), and a `FinalAnswerTool` 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.
* **smolagents spans missing but other spans present.** `SmolagentsInstrumentor().instrument(...)` must run before any `from smolagents 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 for a model your key can call.
* **Other LLM providers.** smolagents supports many backends — `HfApiModel()` (Hugging Face Inference), `LiteLLMModel(...)`, `TransformersModel(...)`, etc. The same `SmolagentsInstrumentor` covers them.
## Resources