> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Gemma 4 generation model (gemini/gemma-4): where to run it

> Gemma 4 is a generation model from Google Gemini. Run it with ComfyUI locally and Comfy Cloud; templates, files, nodes, and pricing.

Model ID: gemini/gemma-4
Provider: gemini
Product: local, cloud, workflows
Input: image, audio, video
License: open-weights

Gemma 4 is a generation model from Google Gemini. It takes image, audio and video as input and produces media. It has open weights, so you can download the files and run it locally in ComfyUI.

## Where you can run it

| Where | Page | Use it when |
| - | - | - |
| ComfyUI (local or Comfy Cloud) | [Gemma 4 in ComfyUI: Multimodal LLM With TextGenerate](https://docs.comfy.org/tutorials/llm/gemma4/gemma4.md) | You follow a step-by-step workflow tutorial |
| Comfy Cloud | [Open `llm_gemma4_text_gen` on Comfy Cloud](https://cloud.comfy.org/?template=llm_gemma4_text_gen) | You want to run a ready-made workflow without installing anything |

## Model files for running it locally

Put each file in the matching folder under `ComfyUI/models/`. These come from the workflow templates below (workflow\_templates at `3c2199b9`).

| File | Folder | Download |
| - | - | - |
| `gemma4_e4b_it_fp8_scaled.safetensors` | `models/text_encoders/` | [Download](https://huggingface.co/Comfy-Org/gemma-4/resolve/main/text_encoders/gemma4_e4b_it_fp8_scaled.safetensors) |

## Workflow templates

| Template | Title | Added |
| - | - | - |
| `llm_gemma4_text_gen` | Gemma4: Text Generation | 2026-05-01 |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.