> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/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.recipes.llm

## Submodules

* **[`nemo_automodel.recipes.llm._dspark_target_build`](/nemo-automodel/nemo_automodel/recipes/llm/_dspark_target_build)**
* **[`nemo_automodel.recipes.llm._spec_train_utils`](/nemo-automodel/nemo_automodel/recipes/llm/_spec_train_utils)**
* **[`nemo_automodel.recipes.llm.benchmark`](/nemo-automodel/nemo_automodel/recipes/llm/benchmark)**
* **[`nemo_automodel.recipes.llm.kd`](/nemo-automodel/nemo_automodel/recipes/llm/kd)**
* **[`nemo_automodel.recipes.llm.peagle_recipe`](/nemo-automodel/nemo_automodel/recipes/llm/peagle_recipe)**
* **[`nemo_automodel.recipes.llm.precompute_dspark_dist`](/nemo-automodel/nemo_automodel/recipes/llm/precompute_dspark_dist)**
* **[`nemo_automodel.recipes.llm.train_dflash`](/nemo-automodel/nemo_automodel/recipes/llm/train_dflash)**
* **[`nemo_automodel.recipes.llm.train_dflash2`](/nemo-automodel/nemo_automodel/recipes/llm/train_dflash2)**
* **[`nemo_automodel.recipes.llm.train_domino`](/nemo-automodel/nemo_automodel/recipes/llm/train_domino)**
* **[`nemo_automodel.recipes.llm.train_dspark`](/nemo-automodel/nemo_automodel/recipes/llm/train_dspark)**
* **[`nemo_automodel.recipes.llm.train_eagle1`](/nemo-automodel/nemo_automodel/recipes/llm/train_eagle1)**
* **[`nemo_automodel.recipes.llm.train_eagle2`](/nemo-automodel/nemo_automodel/recipes/llm/train_eagle2)**
* **[`nemo_automodel.recipes.llm.train_eagle3`](/nemo-automodel/nemo_automodel/recipes/llm/train_eagle3)**
* **[`nemo_automodel.recipes.llm.train_ft`](/nemo-automodel/nemo_automodel/recipes/llm/train_ft)**
* **[`nemo_automodel.recipes.llm.train_jetspec`](/nemo-automodel/nemo_automodel/recipes/llm/train_jetspec)**
* **[`nemo_automodel.recipes.llm.train_seq_cls`](/nemo-automodel/nemo_automodel/recipes/llm/train_seq_cls)**
* **[`nemo_automodel.recipes.llm.train_vispec`](/nemo-automodel/nemo_automodel/recipes/llm/train_vispec)**

## Docs

- [nemo_automodel.recipes.llm.benchmark](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/benchmark.md)
- [nemo_automodel.recipes.llm.kd](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/kd.md)
- [nemo_automodel.recipes.llm.peagle_recipe](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/peagle_recipe.md)
- [nemo_automodel.recipes.llm.precompute_dspark_dist](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/precompute_dspark_dist.md)
- [nemo_automodel.recipes.llm.train_dflash](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_dflash.md)
- [nemo_automodel.recipes.llm.train_dflash2](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_dflash2.md)
- [nemo_automodel.recipes.llm.train_domino](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_domino.md)
- [nemo_automodel.recipes.llm.train_dspark](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_dspark.md)
- [nemo_automodel.recipes.llm.train_eagle1](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_eagle1.md)
- [nemo_automodel.recipes.llm.train_eagle2](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_eagle2.md)
- [nemo_automodel.recipes.llm.train_eagle3](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_eagle3.md)
- [nemo_automodel.recipes.llm.train_ft](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_ft.md)
- [nemo_automodel.recipes.llm.train_jetspec](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_jetspec.md)
- [nemo_automodel.recipes.llm.train_seq_cls](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_seq_cls.md)
- [nemo_automodel.recipes.llm.train_vispec](https://docs.nvidia.com/nemo/automodel/latest/nemo-automodel/nemo_automodel/recipes/llm/train_vispec.md)