> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/llms.txt. # nemo_automodel.components.models.lfm3 ## Submodules * **[`nemo_automodel.components.models.lfm3.config`](/nemo-automodel/nemo_automodel/components/models/lfm3/config)** * **[`nemo_automodel.components.models.lfm3.context_parallel`](/nemo-automodel/nemo_automodel/components/models/lfm3/context_parallel)** * **[`nemo_automodel.components.models.lfm3.model`](/nemo-automodel/nemo_automodel/components/models/lfm3/model)** * **[`nemo_automodel.components.models.lfm3.parallelization`](/nemo-automodel/nemo_automodel/components/models/lfm3/parallelization)** * **[`nemo_automodel.components.models.lfm3.processing`](/nemo-automodel/nemo_automodel/components/models/lfm3/processing)** * **[`nemo_automodel.components.models.lfm3.vl_model`](/nemo-automodel/nemo_automodel/components/models/lfm3/vl_model)** ## Package Contents ### Classes | Name | Description | | --------------------------------------------------------------------------------- | --------------------------------------------------------------------------- | | [`Lfm3ForCausalLM`](#nemo_automodel-components-models-lfm3-model-Lfm3ForCausalLM) | LFM3 with a vocabulary head, loadable from LiquidAI's released checkpoints. | ### API ```python class nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM( config: nemo_automodel.components.models.lfm3.config.Lfm3Config ) ``` **Bases:** [HFCheckpointingMixin](/nemo-automodel/nemo_automodel/components/models/common/hf_checkpointing_mixin#nemo_automodel-components-models-common-hf_checkpointing_mixin-HFCheckpointingMixin), [Lfm3PreTrainedModel](#nemo_automodel-components-models-lfm3-model-Lfm3PreTrainedModel), `GenerationMixin` LFM3 with a vocabulary head, loadable from LiquidAI's released checkpoints. **`_tied_weights_keys`** `= {'lm_head.weight': 'model.embed_tokens.weight'}` --- **`lm_head`** --- **`model`** `= Lfm3Model(config)` --- **`tie_word_embeddings_support`** `TieSupport = TieSupport.BOTH` --- ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.forward( input_ids: torch.LongTensor | None = None, attention_mask: torch.Tensor | dict[str, torch.Tensor | None] | None = None, position_ids: torch.LongTensor | None = None, past_key_values: transformers.cache_utils.Cache | None = None, inputs_embeds: torch.FloatTensor | None = None, labels: torch.LongTensor | None = None, use_cache: bool | None = None, logits_to_keep: int | torch.Tensor = 0, kwargs: transformers.processing_utils.Unpack[transformers.utils.TransformersKwargs] = {} ) -> transformers.modeling_outputs.CausalLMOutputWithPast ``` Predict next-token logits and, given labels, the causal LM loss. **Parameters:** **`input_ids`** `torch.LongTensor | None` — default: None Long Tensor of shape \[batch, sequence]. Exclusive with `inputs_embeds`. --- **`attention_mask`** `torch.Tensor | dict[str, torch.Tensor | None] | None` — default: None Optional padding Tensor of shape \[batch, key_sequence], where `key_sequence` also counts cached tokens. --- **`position_ids`** `torch.LongTensor | None` — default: None Optional long Tensor of shape \[batch, sequence]. --- **`past_key_values`** `Cache | None` — default: None Optional generation cache, updated in place. --- **`inputs_embeds`** `torch.FloatTensor | None` — default: None Tensor of shape \[batch, sequence, hidden]. --- **`labels`** `torch.LongTensor | None` — default: None Optional long Tensor of shape \[batch, sequence]; `-100` is ignored. Shifted internally. --- **`use_cache`** `bool | None` — default: None Whether to create and return a generation cache. --- **`logits_to_keep`** `int | torch.Tensor` — default: 0 Project only the last `logits_to_keep` positions (0 keeps all), or the positions in a long Tensor of shape \[kept]. --- **`**kwargs`** `Unpack[TransformersKwargs]` — default: \{} Hugging Face output-capture, attention and loss arguments. --- **Returns:** `CausalLMOutputWithPast` Output whose `logits` is a Tensor of shape \[batch, kept, vocab] and whose ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.get_input_embeddings() -> torch.nn.Embedding ``` Return the token embedding table. ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.get_output_embeddings() -> torch.nn.Linear ``` Return the vocabulary head. ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.prepare_model_inputs_for_cp( batch: nemo_automodel.components.models.lfm3.context_parallel.CPBatch, num_chunks: int = 1 ) -> dict[str, nemo_automodel.components.distributed.context_parallel.sharder.ContextParallelSharder] ``` Return the sharder that keeps packed documents apart and splits the sequence across CP ranks. **Parameters:** **`batch`** `CPBatch` The recipe batch with token Tensors of shape \[batch, sequence]; it is not read or changed here, the sharder reads it when it shards. --- **`num_chunks`** `int` — default: 1 Unused; LFM3 shards whole rows. --- **Returns:** `dict[str, ContextParallelSharder]` Mapping with the `cp_sharder` built on `shard_documents`. ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.set_input_embeddings( value: torch.nn.Embedding ) -> None ``` Replace the token embedding table and retie the head. ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.tie_weights( kwargs: bool | set[str] | None = {} ) -> None ``` Share the vocabulary head with the token embeddings when the config asks for it. ```python nemo_automodel.components.models.lfm3.model.Lfm3ForCausalLM.validate_packed_attention_backend( attn_implementation: str ) -> None ``` Reject batched packing on backends that give the convolutions no document boundaries. ## Docs - [nemo_automodel.components.models.lfm3.config](https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/config.md) - [nemo_automodel.components.models.lfm3.context_parallel](https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/context_parallel.md) - [nemo_automodel.components.models.lfm3.model](https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/model.md) - [nemo_automodel.components.models.lfm3.parallelization](https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/parallelization.md) - [nemo_automodel.components.models.lfm3.processing](https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/processing.md) - [nemo_automodel.components.models.lfm3.vl_model](https://docs.nvidia.com/nemo/automodel/nightly/nemo-automodel/nemo_automodel/components/models/lfm3/vl_model.md)