Skip to main content
NVIDIA NeMo Agent Toolkit is NVIDIA’s open-source toolkit for building, running, and observing agent workflows. The toolkit ships with a built-in arize_ax OpenTelemetry exporter, so traces — including LLM calls, tool invocations, and full input/output payloads — flow into Arize AX without any extra instrumentation code.
This integration is driven entirely by a YAML configuration file in the NeMo Agent Toolkit repository. There is no standalone Python script to run; you launch your workflow with the toolkit’s nat CLI and the toolkit handles exporter setup.

​
Prerequisites

​
Launch Arize AX

  1. Sign in to your Arize AX account.
  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

Install the OpenTelemetry extra to enable the arize_ax exporter: 
uv pip install "nvidia-nat[opentelemetry]"

​
Configure credentials

export ARIZE_SPACE_ID="<your-space-id>"
export ARIZE_API_KEY="<your-api-key>"
export ARIZE_PROJECT_NAME="simple_calculator"

​
Setup tracing

Add the arize_ax exporter to your workflow’s YAML configuration file: 
general:
  telemetry:
    tracing:
      arize_ax:
        _type: arize_ax
        space_id: ${ARIZE_SPACE_ID}
        api_key: ${ARIZE_API_KEY}
        project: ${ARIZE_PROJECT_NAME:-simple_calculator}
        # Optional: enable for EU data residency
        # use_eu_region: true
This sends OTLP-formatted traces to Arize AX, grouped into the project you specify.

​
Run NeMo Agent Toolkit

From the root of the NeMo Agent Toolkit repository, install the example workflow and run it with your Arize AX configuration: 
uv pip install -e examples/observability/simple_calculator_observability/

nat run \
  --config_file examples/observability/simple_calculator_observability/configs/config-arize-ax.yml \
  --input "What is 2 * 4?"

​
Expected output

Started exporter 'arize_ax'
The result of 2 * 4 is 8.

​
Verify in Arize AX

  1. Open your Arize AX space and select project simple_calculator (or whatever value you set for ARIZE_PROJECT_NAME).
  2. You should see a new trace within ~30 seconds containing the workflow’s root span and nested LLM / tool spans with prompts, responses, and token usage attached.
  3. If no traces appear, see Troubleshooting.

​
Troubleshooting

  • No Started exporter 'arize_ax' log line. The OpenTelemetry extra isn’t installed in the active environment. Re-run uv pip install "nvidia-nat[opentelemetry]" and confirm the YAML _type is exactly arize_ax.
  • No traces in Arize AX. Confirm ARIZE_SPACE_ID and ARIZE_API_KEY are exported in the same shell that runs nat run. The YAML uses ${ARIZE_SPACE_ID} / ${ARIZE_API_KEY} substitution, so unset variables resolve to empty strings and the exporter silently fails to authenticate.
  • EU data residency. Set use_eu_region: true under general.telemetry.tracing.arize_ax to route traces to the EU ingest endpoint.
  • Wrong project in Arize AX. The YAML project: value wins over ARIZE_PROJECT_NAME unless you use the ${ARIZE_PROJECT_NAME:-default} substitution shown above.

​
Resources

Observe NeMo Agent Toolkit with Arize AX

NeMo Agent Toolkit Documentation

NeMo Agent Toolkit GitHub