> This page is for version 0.7.0 · 26.10.
> 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
> - 0.7.0 · 26.10: https://docs.nvidia.com/nemo/automodel/v0.7/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.

# API Reference

> Auto-generated Python API reference for the nemo_automodel library.

Auto-generated reference documentation for the `nemo_automodel` Python package.

This reference is built from docstrings in the [source code](https://github.com/NVIDIA-NeMo/Automodel/tree/r0.7.0/nemo_automodel). Browse the sidebar to explore modules, classes, and functions.

## Key Modules

| Module                                   | Description                                                                                                                                           |
| ---------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------- |
| `nemo_automodel._transformers`           | Hugging Face bridge (`NeMoAutoModelForCausalLM`, `NeMoAutoModelForImageTextToText`, model registry, capability flags, and device-mesh infrastructure) |
| `nemo_automodel._diffusers`              | Diffusion pipeline wrapper (`NeMoAutoDiffusionPipeline`)                                                                                              |
| `nemo_automodel.components.models`       | 27+ model families (LLM, VLM, MoE, retrieval) and their state-dict adapters                                                                           |
| `nemo_automodel.components.datasets`     | LLM, VLM, and diffusion data pipelines                                                                                                                |
| `nemo_automodel.components.distributed`  | FSDP2, HSDP, DDP utilities and device-mesh helpers                                                                                                    |
| `nemo_automodel.components.checkpoint`   | Async DCP and SafeTensors checkpointing                                                                                                               |
| `nemo_automodel.components.quantization` | FP8, QAT, calibration                                                                                                                                 |
| `nemo_automodel.components._peft`        | LoRA, QLoRA adapters                                                                                                                                  |
| `nemo_automodel.components.launcher`     | Interactive, SkyPilot, and NeMo-Run job launchers; Slurm jobs use the repository's `slurm.sub` script                                                 |
| `nemo_automodel.cli`                     | `uv run automodel <config.yaml> [--nproc-per-node N] [--key.subkey=override ...]` entry point                                                         |

## Extension APIs

* [Model Parallelizer API](/nemo/automodel/v0.7/reference/model-parallelizer): Add a model-owned
  parallelization sidecar while keeping topology, dispatch, and default
  TP/CP/EP/FSDP behavior in shared infrastructure.