> 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.minimax_m2.model

## Module Contents

### Classes

| Name                                                                                              | Description                                        |
| ------------------------------------------------------------------------------------------------- | -------------------------------------------------- |
| [`Block`](#nemo_automodel-components-models-minimax_m2-model-Block)                               | MiniMax-M2 transformer block with attention + MoE. |
| [`MiniMaxM2ForCausalLM`](#nemo_automodel-components-models-minimax_m2-model-MiniMaxM2ForCausalLM) | -                                                  |
| [`MiniMaxM2Model`](#nemo_automodel-components-models-minimax_m2-model-MiniMaxM2Model)             | -                                                  |

### Data

[`ModelClass`](#nemo_automodel-components-models-minimax_m2-model-ModelClass)

### API

```python
class nemo_automodel.components.models.minimax_m2.model.Block(
    layer_idx: int,
    config: typing.Any,
    moe_config: nemo_automodel.components.moe.layers.MoEConfig,
    backend: nemo_automodel.components.models.common.BackendConfig
)
```

**Bases:** `Module`

MiniMax-M2 transformer block with attention + MoE.

```python
nemo_automodel.components.models.minimax_m2.model.Block.forward(
    x: torch.Tensor,
    freqs_cis: torch.Tensor,
    attention_mask: torch.Tensor | None = None,
    padding_mask: torch.Tensor | None = None,
    attn_kwargs: typing.Any = {}
) -> torch.Tensor
```

```python
nemo_automodel.components.models.minimax_m2.model.Block.init_weights(
    buffer_device: torch.device
)
```

```python
class nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM(
    config: typing.Any,
    moe_config: nemo_automodel.components.moe.layers.MoEConfig | None = None,
    backend: nemo_automodel.components.models.common.BackendConfig | None = None,
    kwargs = {}
)
```

**Bases:** [HFCheckpointingMixin](/nemo-automodel/nemo_automodel/components/models/common/hf_checkpointing_mixin#nemo_automodel-components-models-common-hf_checkpointing_mixin-HFCheckpointingMixin), `Module`, [MoEFSDPSyncMixin](/nemo-automodel/nemo_automodel/components/moe/fsdp_mixin#nemo_automodel-components-moe-fsdp_mixin-MoEFSDPSyncMixin)

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.forward(
    input_ids: torch.Tensor,
    position_ids: torch.Tensor | None = None,
    attention_mask: torch.Tensor | None = None,
    padding_mask: torch.Tensor | None = None,
    logits_to_keep: typing.Union[int, torch.Tensor] = 0,
    output_hidden_states: typing.Optional[bool] = None,
    attn_kwargs: typing.Any = {}
) -> transformers.modeling_outputs.CausalLMOutputWithPast
```

Forward pass returning :class:`~transformers.modeling_outputs.CausalLMOutputWithPast`.

**Parameters:**

Input token IDs. BSHD: `[B, S]`; THD: `[1, T]` (squeezed internally).

Optional position indices.

Optional attention mask.

Optional padding mask.

If 0, compute logits for all positions (training default);
otherwise only compute logits for the last `logits_to_keep` token positions.

Whether to return the final hidden states on the output.

Additional arguments forwarded to the base model (e.g. qkv\_format).

**Returns:** `CausalLMOutputWithPast`

class:`~transformers.modeling_outputs.CausalLMOutputWithPast` with `logits`

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.from_config(
    config: typing.Any,
    moe_config: nemo_automodel.components.moe.layers.MoEConfig | None = None,
    backend: nemo_automodel.components.models.common.BackendConfig | None = None,
    kwargs = {}
)
```

classmethod

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.from_pretrained(
    pretrained_model_name_or_path: str,
    model_args = (),
    kwargs = {}
)
```

classmethod

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.get_input_embeddings()
```

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.get_output_embeddings()
```

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.initialize_weights(
    buffer_device: torch.device | None = None,
    dtype: torch.dtype = torch.bfloat16
) -> None
```

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.set_input_embeddings(
    value
)
```

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2ForCausalLM.set_output_embeddings(
    new_embeddings
)
```

```python
class nemo_automodel.components.models.minimax_m2.model.MiniMaxM2Model(
    config: typing.Any,
    backend: nemo_automodel.components.models.common.BackendConfig,
    moe_config: nemo_automodel.components.moe.layers.MoEConfig | None = None,
    moe_overrides: dict | None = None
)
```

**Bases:** `Module`

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2Model.forward(
    input_ids: torch.Tensor,
    position_ids: torch.Tensor | None = None,
    attention_mask: torch.Tensor | None = None,
    padding_mask: torch.Tensor | None = None,
    attn_kwargs: typing.Any = {}
) -> torch.Tensor
```

```python
nemo_automodel.components.models.minimax_m2.model.MiniMaxM2Model.init_weights(
    buffer_device: torch.device | None = None
) -> None
```

```python
nemo_automodel.components.models.minimax_m2.model.ModelClass = MiniMaxM2ForCausalLM
```