> 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.

# FunctionGemma

> Supported FunctionGemma checkpoint and NeMo AutoModel function-calling recipe.

[FunctionGemma](https://huggingface.co/google/functiongemma-270m-it) is a compact Gemma model specialized for function calling.

|                  |                                         |
| ---------------- | --------------------------------------- |
| **Task**         | Function Calling / Text Generation      |
| **Architecture** | `Gemma3ForCausalLM`                     |
| **Parameters**   | 270M                                    |
| **HF Org**       | [google](https://huggingface.co/google) |

## Example Model and Recipe

| Model                                                                                 | Recipe                                                                                                                              |
| ------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------- |
| [`google/functiongemma-270m-it`](https://huggingface.co/google/functiongemma-270m-it) | [xLAM function-calling SFT](https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm_finetune/gemma/functiongemma_xlam.yaml) |

```bash
uv run automodel --nproc-per-node=8 examples/llm_finetune/gemma/functiongemma_xlam.yaml
```

See the [FunctionGemma Fine-Tuning Guide](/recipes-e2e-examples/function-calling).