CLI Reference | NVIDIA NeMo Platform

This reference covers the nemo anonymizer commands exposed by the Anonymizer plugin. For end-to-end walkthroughs, see the tutorials.

Command Surface

Command	Source	Description
`nemo anonymizer validate`	Manual Typer command	Validate an `AnonymizerConfig` (and optional `model_configs`).
`nemo anonymizer preview run`	Generated from `NemoFunction`	Local streaming preview.
`nemo anonymizer preview submit`	Generated from `NemoFunction`	Remote streaming preview against the plugin service.
`nemo anonymizer run run`	Generated from `NemoJob`	Local job execution in the CLI process.
`nemo anonymizer run submit`	Generated from `NemoJob`	Submit an `anonymizer.run` job to the NeMo Platform Jobs worker.
`nemo anonymizer run explain`	Generated from `NemoJob`	Print the job key, submit endpoint, and JSON schemas.

`nemo anonymizer validate`

Validate an AnonymizerConfig YAML file against the model selection. Useful for catching Substitute-without-replacement_generator and similar misconfigurations before submitting a request.

$ nemo anonymizer validate \
>   --config /tmp/anonymizer-config.yaml \
>   [--model-configs /tmp/anonymizer-model-configs.yaml]

Flag	Required	Description
`--config`	yes	Path to the `AnonymizerConfig` YAML.
`--model-configs`	no	Optional path to a model-configs YAML to validate alongside the config (same shape the library expects).

The command does not accept data.source. Input-source validation happens during preview or run.

`nemo anonymizer preview`

Both preview run and preview submit take a spec file matching PreviewRequest.

$ nemo anonymizer preview run \
>   --spec-file /tmp/anonymizer-preview.yaml \
>   --workspace "${NMP_WORKSPACE:-default}"
$ 
$ nemo anonymizer preview submit \
>   --spec-file /tmp/anonymizer-preview.yaml \
>   --workspace "${NMP_WORKSPACE:-default}" \
>   --base-url "${NMP_BASE_URL:-http://localhost:8080}"

Flag	Description
`--spec-file`	Path to the `PreviewRequest` YAML.
`--workspace`	NeMo Platform workspace. Used to resolve fileset references and Inference Gateway providers.
`--base-url`	Override the platform base URL (typically auto-populated from the CLI config).

Preview source kinds

Form	`preview run`	`preview submit`
Local path (`/tmp/input.csv`)	yes	no
HTTP(S) URL (`https://.../input.csv`)	yes	yes
Fileset reference (`fs#path`)	yes	yes

Preview output

preview streams newline-delimited JSON frames to stdout. Filter with jq:

$ nemo anonymizer preview run --spec-file /tmp/anonymizer-preview.yaml > /tmp/preview.ndjson
$ 
$ jq -R 'fromjson? | select(.kind == "preview_dataset") | .records' /tmp/preview.ndjson

Frame kinds: log, preview_dataset, trace_dataset, failed_records, heartbeat, done, error. See the preview tutorial for details.

`nemo anonymizer run`

$ nemo anonymizer run run --spec-file /tmp/anonymizer-run.yaml
$ 
$ nemo anonymizer run submit \
>   --spec-file /tmp/anonymizer-run.yaml \
>   --workspace "${NMP_WORKSPACE:-default}" \
>   --base-url "${NMP_BASE_URL:-http://localhost:8080}"
$ 
$ nemo anonymizer run explain

Flag	Description
`--spec-file`	Path to the `AnonymizerRequest` YAML.
`--workspace`	Workspace used for fileset resolution, Inference Gateway providers, and job placement.

Run source kinds

Form	`run run`	`run submit`
Local path (`/tmp/input.csv`)	yes	no
HTTP(S) URL (`https://.../input.csv`)	yes	yes
Fileset reference (`fs#path`)	yes	yes

Run output

run run prints {"exit_code": 0} on success. The local job results manager logs the artifact directory to stderr:

Saved result 'artifacts' to file:///.../persistent/results/artifacts

The artifact directory contains:

File	Description
`dataset.parquet`	Anonymized output.
`trace.parquet`	Detection trace.
`metadata.json`	Run metadata (includes original text column).
`failed_records.json`	Per-record failures. Only written when records failed.

run submit submits an anonymizer.run job to the Jobs service and prints the assigned job name and submit endpoint:

  |-- job name: anonymizer-run-2026-05-12-abc123
  |-- submit endpoint: /apis/anonymizer/v2/workspaces/default/jobs/run
{"name": "anonymizer-run-2026-05-12-abc123", ...}

Track and pull artifacts using either the standard nemo jobs ... commands or the Python SDK:

$ nemo jobs get-status anonymizer-run-2026-05-12-abc123 --workspace "${NMP_WORKSPACE:-default}"
$ nemo jobs get-logs anonymizer-run-2026-05-12-abc123 --workspace "${NMP_WORKSPACE:-default}"

1 job = sdk.anonymizer.get_job_resource("anonymizer-run-2026-05-12-abc123")
2 job.wait_until_done()
3 results = job.download_artifacts()
4 dataset = results.load_dataset()

See SDK Resources for the full AnonymizerJobResource / AnonymizerJobResults surface.

Compared to run run, run submit rejects local file paths in data.source (use a fileset reference or http(s) URL) and requires explicit model_configs because the job runs outside the CLI process.

Spec File Reference

Both preview and run specs use the shared AnonymizerRequest shape:

Field	Type	Required	Notes
`config`	`AnonymizerConfig`	yes	Library config. See the library docs.
`data.source`	string	yes	Local path, `http(s)` URL, or fileset reference.
`data.text_column`	string	no	Defaults to `text`.
`data.id_column`	string	no	Optional record identifier column.
`data.data_summary`	string	no	Optional short description of the data.
`model_configs`	list of Data Designer `ModelConfig`	depends	Required for `preview submit` and `run submit`; optional for `preview run` and `run run`.
`selected_models`	object with `detection` / `replace` / `rewrite`	no	Role overrides on top of bundled defaults. Requires `model_configs`.

Preview-only:

Field	Type	Required	Notes
`num_records`	int ≥ 1	no	Defaults to 10. Capped by the service’s `preview_num_records.max`.

Fileset Reference Forms

The data.source field accepts three fileset forms; the workspace and fileset must already exist:

fileset://<workspace>/<fileset>#<path>
<workspace>/<fileset>#<path>
<fileset>#<path>

The #<path> fragment must resolve to a single .csv or .parquet file. The plugin downloads the file before constructing the Anonymizer library input.