Basic Audit Target#

When you run an audit job in NVIDIA NeMo Auditor, you create a separate audit target and audit configuration for the job.

The target specifies the model name, model type, and free-form key-value pairs for model-specific inference options. For more information, refer to Schema for Audit Targets.

The following target references the NVIDIA Nemotron Nano 12B V2 VL model from build.nvidia.com.

Set the NMP_BASE_URL environment variable to the NeMo Auditor service endpoint. Refer to Accessing the Microservice for more information.

With NeMo Platform running, follow the model provider instructions to set up a model provider that points at integrate.api.nvidia.com and the model to audit. Create the provider in the default workspace and name it build.

Python SDK

    import os
    from nemo_platform import NeMoPlatform

    client = NeMoPlatform(
    base_url=os.environ.get("NMP_BASE_URL", "http://localhost:8080"),
    workspace="default",
)

    target = client.audit.targets.create(
        workspace="default",
        name="demo-basic-target",
        type="nim.NVOpenAIChat",
        model="nvidia/nemotron-nano-12b-v2-vl",
        options={
            "nim": {
                "skip_seq_start": "<think>",
                "skip_seq_end": "</think>",
                "max_tokens": 3200,
                "nmp_uri_spec": {"inference_gateway": { "workspace": "default", "provider": "build" }}
            }
        }
    )

    print(target.model_dump_json(indent=2))

CLI

	nmp audit targets create --workspace "default" \
	  --model "nvidia/nemotron-nano-12b-v2-vl" \
	  --name "demo-simple-target" \
	  --type "nim" \
	  --options '{"nim": { "skip_seq_start": "<think>", "skip_seq_end": "</think>", "max_tokens": 4000,
              "nmp_uri_spec": {"inference_gateway": { "workspace": "default", "provider": "build" }}}}'

Basic Audit Target#

Next Steps#