> This page is for version Nightly (default).
> For other versions, use one of these documentation indexes:
> - Nightly (default): https://docs.nvidia.com/nemo/automodel/nightly/llms.txt
> - Latest: https://docs.nvidia.com/nemo/automodel/latest/llms.txt
> - 0.5.0 · 26.06: https://docs.nvidia.com/nemo/automodel/v0.5/llms.txt
> - 0.4.0 · 26.04: https://docs.nvidia.com/nemo/automodel/v0.4/llms.txt

> 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.beam

## Submodules

* **[`nemo_automodel.components.models.beam.layers`](/nemo-automodel/nemo_automodel/components/models/beam/layers)**
* **[`nemo_automodel.components.models.beam.model`](/nemo-automodel/nemo_automodel/components/models/beam/model)**
* **[`nemo_automodel.components.models.beam.parallelization`](/nemo-automodel/nemo_automodel/components/models/beam/parallelization)**
* **[`nemo_automodel.components.models.beam.state_dict_adapter`](/nemo-automodel/nemo_automodel/components/models/beam/state_dict_adapter)**
* **[`nemo_automodel.components.models.beam.tokenization`](/nemo-automodel/nemo_automodel/components/models/beam/tokenization)**

## Package Contents

### Classes

| Name                                                                                                     | Description                                                                           |
| -------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
| [`BeamForCausalLM`](#nemo_automodel-components-models-beam-model-BeamForCausalLM)                        | Beam causal LM on Automodel's MoE stack (see module docstring).                       |
| [`BeamStateDictAdapter`](#nemo_automodel-components-models-beam-state_dict_adapter-BeamStateDictAdapter) | Converts between Hugging Face Beam checkpoints and the grouped-experts native format. |

### Data

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

### API

```python
class nemo_automodel.components.models.beam.model.BeamForCausalLM(
    config: transformers.models.beam.BeamConfig,
    moe_config: nemo_automodel.components.moe.config.MoEConfig | None = None,
    backend: nemo_automodel.components.models.common.BackendConfig | None = None,
    kwargs: typing.Any = {}
)
```

**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)

Beam causal LM on Automodel's MoE stack (see module docstring).

**`_keep_in_fp32_modules_strict`** `list[str] = ['mlp.gate.weight', 'e_score_correction_bias']`

---

**`_pp_keep_self_forward`** `bool = True`

---

**`backend`**

---

**`final_logit_softcapping`** `= getattr(config, 'final_logit_softcapping', None)`

---

**`lm_head`**

---

**`model`**

---

**`state_dict_adapter`**

---

**`tie_word_embeddings_support`** `TieSupport = TieSupport.UNTIED_ONLY`

---

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.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: int | torch.Tensor = 0,
    output_hidden_states: bool | None = None,
    attn_kwargs: typing.Any = {}
) -> transformers.modeling_outputs.CausalLMOutputWithPast
```

Return soft-capped logits or pipeline-stage hidden states.

**Parameters:**

**`input_ids`** `torch.Tensor`

Integer tensor \[batch, sequence] or packed \[tokens]; later PP stages
receive floating hidden states with an extra final hidden axis. Positions are local under CP.

---

**`position_ids`** `torch.Tensor | None` — default: None

Optional integer tensor \[batch, sequence] or \[tokens], absolute within each document.

---

**`attention_mask`** `torch.Tensor | None` — default: None

Optional \[batch, sequence] bool/0-1 mask; True/1 means valid. None on flex.

---

**`padding_mask`** `torch.Tensor | None` — default: None

Optional bool tensor matching the position axes; True skips routing.

---

**`logits_to_keep`** `int | torch.Tensor` — default: 0

Zero for all positions, a positive final-position count, or integer
tensor \[selected_positions] indexing the sequence axis.

---

**`output_hidden_states`** `bool | None` — default: None

Whether to return the final hidden states.

---

**`**attn_kwargs`** `Any` — default: \{}

Backend metadata: qkv_format (bshd/thd), integer cu_seqlens and
cu_seqlens_padded \[documents + 1], scalar max_seqlen, cp_size and cp_rank.

---

**Returns:** `CausalLMOutputWithPast`

logits \[batch, selected_sequence, vocab] or packed

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.from_config(
    config: transformers.models.beam.BeamConfig,
    moe_config: nemo_automodel.components.moe.config.MoEConfig | None = None,
    backend: nemo_automodel.components.models.common.BackendConfig | None = None,
    kwargs: typing.Any = {}
) -> nemo_automodel.components.models.beam.model.BeamForCausalLM
```

classmethod

Construct the native model from an HF BeamConfig and optional backend/MoE overrides.

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.from_pretrained(
    pretrained_model_name_or_path: str,
    model_args: typing.Any = (),
    kwargs: typing.Any = {}
) -> nemo_automodel.components.models.beam.model.BeamForCausalLM
```

classmethod

Construct from the checkpoint's config; framework checkpoint loading supplies the weights.

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.get_input_embeddings() -> torch.nn.Module | None
```

Return this stage's embedding module, or None when another stage owns it.

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.get_output_embeddings() -> torch.nn.Module | None
```

Return this stage's LM head, or None when another stage owns it.

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

Initialize weights and cast to dtype while preserving original FP32 router values.

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.set_input_embeddings(
    value: torch.nn.Module | None
) -> None
```

Replace this stage's embedding module; None marks a later pipeline stage.

```python
nemo_automodel.components.models.beam.model.BeamForCausalLM.set_output_embeddings(
    new_embeddings: torch.nn.Module | None
) -> None
```

Replace this stage's LM head; None marks a non-final pipeline stage.

```python
class nemo_automodel.components.models.beam.state_dict_adapter.BeamStateDictAdapter()
```

**Bases:** [Qwen3MoeStateDictAdapter](/nemo-automodel/nemo_automodel/components/models/qwen3_moe/state_dict_adapter#nemo_automodel-components-models-qwen3_moe-state_dict_adapter-Qwen3MoeStateDictAdapter)

Converts between Hugging Face Beam checkpoints and the grouped-experts native format.

Beam stores its experts exactly like Qwen3-MoE, so the Qwen3-MoE conversion (including fused PEFT v5
ParamWrapper LoRA export/import and low-memory DCP loading) applies unchanged. The native module tree
mirrors the HF one, so every tensor outside the experts keeps its HF name and passes through unchanged
(attention projections and `gate_bias`, q/k norms, the four per-layer norms, the dense MLP,
`mlp.gate.weight` and the `mlp.gate.e_score_correction_bias` buffer, embeddings, final norm,
`lm_head`). Experts differ:

HF checkpoint (per expert, the `qwen2_moe` conversion mapping)::

model.layers.\{L}.mlp.experts.\{E}.gate_proj.weight   \[moe_inter_dim, hidden]
model.layers.\{L}.mlp.experts.\{E}.up_proj.weight     \[moe_inter_dim, hidden]
model.layers.\{L}.mlp.experts.\{E}.down_proj.weight   \[hidden, moe_inter_dim]

Native (grouped)::

model.layers.\{L}.mlp.experts.gate_and_up_projs      \[n_experts, hidden, 2 \* moe_inter_dim]
model.layers.\{L}.mlp.experts.down_projs             \[n_experts, moe_inter_dim, hidden]

`to_hf` therefore writes the per-expert layout that `transformers` itself produces with
`save_pretrained`, not the stacked in-memory layout.

```python
nemo_automodel.components.models.beam.model.ModelClass = BeamForCausalLM
```