chore: Bump data-designer and anonymizer library versions

.github/wheel-constraints/nemo-platform-services.txt

@@ -17,7 +17,7 @@ boto3==1.40.61
 botocore==1.40.61
 clickhouse-connect==0.15.1
 cloudpickle==3.1.2
-data-designer==0.6.1
+data-designer==0.7.0
 datasets==4.3.0
 distro==1.9.0
 docker==7.2.0
@@ -42,7 +42,7 @@ langchain-openai==1.3.5
 langchain==1.3.13
 lark==1.3.1
 litellm<1.92  # 1.92.0 native build has no py3.14 wheel
-nemo-anonymizer==0.2.1
+nemo-anonymizer==0.3.0
 nemo-safe-synthesizer==0.1.7
 nemoguardrails==0.23.0
 ngcsdk==4.20.1

docs/data-designer/index.mdx

@@ -10,7 +10,7 @@ Data Designer on NeMo Platform enables high-quality synthetic data generation th
 
 Data Designer is a framework for orchestrating complex synthetic data generation workflows. It coordinates LLM calls, manages dependencies between data fields, handles batching and parallelization, and validates generated data against specifications.
 
-The plugin is built on the open-source [NVIDIA NeMo Data Designer library](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/getting-started/welcome) ([GitHub](https://github.com/NVIDIA-NeMo/DataDesigner)). The library provides the configuration and generation engine; the plugin provides CLI, SDK, Data Designer API, Jobs, Files API, Secrets API, and Inference Gateway API integration.
+The plugin is built on the open-source [NVIDIA NeMo Data Designer library](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/getting-started/welcome) ([GitHub](https://github.com/NVIDIA-NeMo/DataDesigner)). The library provides the configuration and generation engine; the plugin provides CLI, SDK, Data Designer API, Jobs, Files API, Secrets API, and Inference Gateway API integration.
 
 ## How It Works
 
@@ -46,7 +46,7 @@ config_builder.add_column(dd.LLMTextColumnConfig(...))
 
 Configuration code describes the dataset schema, columns, dependencies, constraints, seed data, processors, profilers, and inference settings.
 
-**Learn more**: See the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/getting-started/welcome) for comprehensive guides on column types, samplers, constraints, and advanced features.
+**Learn more**: See the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/getting-started/welcome) for comprehensive guides on column types, samplers, constraints, and advanced features.
 
 ### 2. Choose Where to Execute
 
@@ -99,7 +99,7 @@ Learn through examples: basics, seeding, and more.
 Move configurations between local CLI and NeMo Services execution.
 
 </Card>
-<Card title="Library Documentation" href="https://docs.nvidia.com/nemo/datadesigner/v0.6.1/getting-started/welcome">
+<Card title="Library Documentation" href="https://docs.nvidia.com/nemo/datadesigner/v0.7.0/getting-started/welcome">
 
 Comprehensive guides on column types, constraints, and advanced features.
 

docs/data-designer/migration.mdx

@@ -83,4 +83,4 @@ Before switching execution modes, verify:
 - **Execution Modes:** See [Execution Modes](/documentation/design-synthetic-data/execution-modes) for the conceptual model.
 - **CLI:** See [Data Designer CLI](/documentation/design-synthetic-data/cli) for `run`, `submit`, and persona commands.
 - **Tutorials:** Follow the [tutorials](/documentation/design-synthetic-data/tutorials) for hands-on examples.
-- **Library Docs:** Refer to the [open-source library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/getting-started/welcome) for configuration details.
+- **Library Docs:** Refer to the [open-source library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/getting-started/welcome) for configuration details.

docs/data-designer/tutorials/basics.mdx

@@ -6,7 +6,7 @@ description: ""
 
 This tutorial demonstrates the fundamentals of Data Designer by generating a product review dataset.
 
-For more detail about column behavior, see the [open-source library's version](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/tutorials/the-basics) of this tutorial.
+For more detail about column behavior, see the [open-source library's version](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/tutorials/the-basics) of this tutorial.
 
 ## Prerequisites
 
@@ -47,7 +47,7 @@ config_builder = dd.DataDesignerConfigBuilder(model_configs)
 
 ### Add Columns
 
-Define the columns for your dataset. The [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/tutorials/the-basics) explains these column types in detail.
+Define the columns for your dataset. The [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/tutorials/the-basics) explains these column types in detail.
 
 
 ```python
@@ -305,5 +305,5 @@ When you use CLI `submit` or the SDK today:
 
 - **Seed data:** Learn how to use external datasets in the [seeding tutorial](/documentation/design-synthetic-data/tutorials/seeding-with-external-datasets)
 - **Execution modes:** Learn more about local and NeMo Services execution in [Execution Modes](/documentation/design-synthetic-data/execution-modes)
-- **Column types:** Explore all available column types in the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/concepts/columns)
-- **Advanced features:** Learn about [processors](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/concepts/processors) and [validation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/concepts/validators)
+- **Column types:** Explore all available column types in the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/concepts/columns)
+- **Advanced features:** Learn about [processors](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/concepts/processors) and [validation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/concepts/validators)

docs/data-designer/tutorials/index.mdx

@@ -18,7 +18,7 @@ Data Designer separates **configuration** (building dataset schemas) from **exec
 
 **Part 1: Build Configs (Library)**
 
-Use `data_designer.config` to define your dataset. See the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/getting-started/welcome) for comprehensive guides on column types, constraints, and processors.
+Use `data_designer.config` to define your dataset. See the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/getting-started/welcome) for comprehensive guides on column types, constraints, and processors.
 
 ```python
 import data_designer.config as dd

docs/data-designer/tutorials/seeding.mdx

@@ -6,7 +6,7 @@ description: ""
 
 This tutorial demonstrates how to use external datasets as seed data for synthetic data generation in Data Designer.
 
-For more detail about seed dataset behavior, see the [open-source library's version](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/tutorials/seeding-with-an-external-dataset) of this tutorial.
+For more detail about seed dataset behavior, see the [open-source library's version](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/tutorials/seeding-with-an-external-dataset) of this tutorial.
 
 ## Seed Sources by Execution Mode
 
@@ -344,5 +344,5 @@ When you configure a seed dataset:
 ## Next Steps
 
 - **Execution modes:** Learn more about local and NeMo Services execution in [Execution Modes](/documentation/design-synthetic-data/execution-modes)
-- **Column types:** Explore all available column types in the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/concepts/columns)
-- **Processors:** Transform your data with processors in the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.6.1/concepts/processors)
+- **Column types:** Explore all available column types in the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/concepts/columns)
+- **Processors:** Transform your data with processors in the [library documentation](https://docs.nvidia.com/nemo/datadesigner/v0.7.0/concepts/processors)

packages/data_designer_nemo/pyproject.toml

@@ -6,7 +6,7 @@ requires-python = ">=3.11,<3.15"
 authors = [{ name = "NVIDIA", email = "nemo@nvidia.com" }]
 dependencies = [
     "nemo-platform",
-    "data-designer==0.6.1",
+    "data-designer==0.7.0",
     "anyio>=4.0",
     "duckdb",
     "pydantic>=2",

packages/data_designer_nemo/src/data_designer_nemo/model_provider.py

@@ -54,7 +54,6 @@ def make_null_registry() -> ModelProviderRegistry:
     # is semantically valid. The library requires a non-empty ModelProviderRegistry, so in this scenario
     # we can provide this dummy null registry.
     return ModelProviderRegistry(
-        default=_NO_OP,
         providers=[make_noop_provider()],
     )
 
@@ -74,12 +73,6 @@ async def make_local_first_model_provider_registry(
     if len(model_configs) == 0:
         return None
 
-    missing_providers = [model_config for model_config in model_configs if model_config.provider is None]
-    if len(missing_providers) > 0:
-        raise NDDInvalidConfigError(
-            f"Error: following model configs do not have an explicit provider defined: {missing_providers}"
-        )
-
     logger.info("Building model provider registry. First checking locally-defined providers.")
 
     local_registry = _make_local_model_provider_registry()
@@ -118,10 +111,7 @@ async def make_local_first_model_provider_registry(
 
     all_providers = local_providers + igw_registry.providers
 
-    return ModelProviderRegistry(
-        default=all_providers[0].name,
-        providers=all_providers,
-    )
+    return ModelProviderRegistry(providers=all_providers)
 
 
 async def _get_igw_model_provider_registry(
@@ -249,11 +239,7 @@ def get_model_provider_registry(self) -> ModelProviderRegistry | None:
 
         if len(self.providers.values()) > 0:
             registry_providers = [providers_tuple[0] for providers_tuple in self.providers.values()]
-            default = registry_providers[0].name
-            return ModelProviderRegistry(
-                default=default,
-                providers=registry_providers,
-            )
+            return ModelProviderRegistry(providers=registry_providers)
 
 
 async def make_model_provider_registry(

packages/data_designer_nemo/tests/unit/test_fileset_file_seed_reader.py

@@ -6,7 +6,6 @@
 import pytest
 from data_designer_nemo.fileset_file_seed_reader import FilesetFileSeedReader, workspace_cvar
 from data_designer_nemo.fileset_file_seed_source import FilesetFileSeedSource
-from nemo_platform_plugin.files.client import FilesClient
 
 
 def test_dataset_uri_with_workspace() -> None:
@@ -40,18 +39,8 @@ def test_create_duckdb_connection_requires_injected_sdk() -> None:
 def test_create_duckdb_connection_uses_injected_sdk() -> None:
     sdk = Mock()
     conn = Mock()
-    mock_files_client = Mock()
-
-    with (
-        patch("data_designer_nemo.fileset_file_seed_reader.duckdb.connect", return_value=conn),
-        patch("data_designer_nemo.fileset_file_seed_reader.FilesetFileSystem") as fileset_file_system,
-        patch(
-            "data_designer_nemo.fileset_file_seed_reader.client_from_platform",
-            return_value=mock_files_client,
-        ) as mock_adapter,
-    ):
+
+    with patch("data_designer_nemo.fileset_file_seed_reader.duckdb.connect", return_value=conn):
         assert FilesetFileSeedReader(sdk).create_duckdb_connection() is conn
 
-    mock_adapter.assert_called_once_with(sdk, FilesClient)
-    fileset_file_system.assert_called_once_with(client=mock_files_client)
-    conn.register_filesystem.assert_called_once_with(fileset_file_system.return_value)
+    conn.register_filesystem.assert_called_once_with(sdk.files.fsspec)

packages/data_designer_nemo/tests/unit/test_model_configs.py

@@ -11,6 +11,7 @@ def _make_model_config(alias: str) -> dd.ModelConfig:
     return dd.ModelConfig(
         alias=alias,
         model="nvidia/nemotron-3",
+        provider="default/nvidia",
     )
 
 

packages/nemo_platform/pyproject.toml

@@ -87,7 +87,7 @@ core-service = [
 
 # Generated from [tool.bundle-package]; do not edit by hand.
 data-designer-nemo = [
-  "data-designer==0.6.1",
+  "data-designer==0.7.0",
   "anyio>=4.0",
   "duckdb",
   "pydantic>=2",
@@ -242,8 +242,8 @@ nemo-agents-plugin = [
 # Generated from [tool.bundle-package]; do not edit by hand.
 nemo-anonymizer-plugin = [
   "nemo-platform-plugin",
-  "nemo-anonymizer>=0.2.1",
-  "data-designer==0.6.1",
+  "nemo-anonymizer>=0.3.0",
+  "data-designer==0.7.0",
   "data-designer-nemo",
   "httpx>=0.27",
   "fastapi",
@@ -264,7 +264,7 @@ nemo-auditor-plugin = [
 # Generated from [tool.bundle-package]; do not edit by hand.
 nemo-data-designer-plugin = [
   "nemo-platform-plugin",
-  "data-designer==0.6.1",
+  "data-designer==0.7.0",
   "data-designer-nemo",
   "httpx>=0.27",
   "fastapi",

plugins/nemo-anonymizer/pyproject.toml

@@ -7,8 +7,8 @@ authors = [{ name = "NVIDIA", email = "nemo@nvidia.com" }]
 dependencies = [
     "nemo-platform-plugin",
     "nemo-platform",
-    "nemo-anonymizer>=0.2.1",
-    "data-designer==0.6.1",
+    "nemo-anonymizer>=0.3.0",
+    "data-designer==0.7.0",
     "data-designer-nemo",
     "httpx>=0.27",
     "fastapi",

plugins/nemo-anonymizer/tests/unit/test_run_job.py

@@ -143,6 +143,7 @@ async def test_run_local_allows_missing_model_configs(
         is_local=True,
     )
 
+    assert isinstance(step_config, AnonymizerStepConfig)
     assert step_config.model_configs_yaml == ""
     assert step_config.dd_model_providers == []
     round_tripped = AnonymizerStepConfig.model_validate(step_config.model_dump())
@@ -157,7 +158,6 @@ async def test_run_local_model_configs_uses_injected_async_sdk(
     csv = tmp_path / "input.csv"
     csv.write_text("text\nhello\n")
     local_first_registry = ModelProviderRegistry(
-        default="local-provider",
         providers=[NDDModelProvider(name="local-provider", endpoint="http://localhost:8000")],
     )
     local_first_lookup = AsyncMock(return_value=local_first_registry)
@@ -179,7 +179,9 @@ async def test_run_local_model_configs_uses_injected_async_sdk(
     )
 
     local_first_lookup.assert_awaited_once()
+    assert local_first_lookup.await_args is not None
     assert local_first_lookup.await_args.kwargs["sdk"] is async_sdk
+    assert isinstance(step_config, AnonymizerStepConfig)
     assert len(step_config.dd_model_providers) == 1
     assert step_config.dd_model_providers[0]["name"] == "local-provider"