> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/automodel/_mcp/server.

# nemo_automodel.components.models.deepseek_v4.parallelization

Model-owned distributed parallelization for DeepSeek-V4.

## Module Contents

### Classes

| Name                                                                                                                       | Description                                                         |
| -------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------- |
| [`DeepseekV4ModelParallelizer`](#nemo_automodel-components-models-deepseek_v4-parallelization-DeepseekV4ModelParallelizer) | Keep DeepSeek-V4 reference-sensitive parameters in fp32 FSDP units. |

### Data

[`PARALLELIZER`](#nemo_automodel-components-models-deepseek_v4-parallelization-PARALLELIZER)

[`__all__`](#nemo_automodel-components-models-deepseek_v4-parallelization-__all__)

### API

```python
class nemo_automodel.components.models.deepseek_v4.parallelization.DeepseekV4ModelParallelizer()
```

**Bases:** `ModelParallelizer`

Keep DeepSeek-V4 reference-sensitive parameters in fp32 FSDP units.

```python
nemo_automodel.components.models.deepseek_v4.parallelization.DeepseekV4ModelParallelizer._fully_shard_module(
    module,
    kwargs = {}
)
```

```python
nemo_automodel.components.models.deepseek_v4.parallelization.PARALLELIZER = DeepseekV4ModelParallelizer()
```

```python
nemo_automodel.components.models.deepseek_v4.parallelization.__all__ = ['PARALLELIZER']
```