> 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.

# Gemma

> Supported text-only Gemma model generations and NeMo AutoModel recipes.

[Google's Gemma](https://ai.google.dev/gemma) is a family of open-weight models
built on the same research and technology as Gemini. The LLM coverage pages
include the text-only Gemma 1 and Gemma 2 generations plus FunctionGemma.

## Model Generations

| Generation    | LLM coverage                                                                   |
| ------------- | ------------------------------------------------------------------------------ |
| Gemma 1       | [Models and recipes](/model-coverage/large-language-models/google/gemma-1)     |
| Gemma 2       | [Models and recipes](/model-coverage/large-language-models/google/gemma-2)     |
| FunctionGemma | [Model and recipe](/model-coverage/large-language-models/google/functiongemma) |

Gemma 3 and Gemma 4 are unified multimodal generations. See the
[Gemma 3 / Gemma 3n](/model-coverage/vision-language-models/google/gemma-3-vl-gemma-3n)
and [Gemma 4](/model-coverage/vision-language-models/google/gemma-4) coverage
pages for those models and recipes.