# Gemma 4

> Open Gemma 4 model evaluated in local AI tests on Macs

- Category: Agents & dev (`agents`)
- Official: https://ai.google.dev/gemma
- Wunderlust: https://wunderlust.xyz/r/gemma-4
- Image: https://wunderlust.xyz/media/gemma-4-poster.jpg
- Video preview: https://wunderlust.xyz/media/gemma-4.webm
- Added: 2026-11-05

## What it is

Language model tested on Macs to measure memory, prefill time and generation.

## What it’s for

For deciding whether the model runs well on your hardware, based on long-context and coding measurements.

## Benefits

- Memory and speed measurements for long context and coding

## Caveats

- Description based on third-party accounts
- Numbers and conclusions about buying hardware are the testing author's claims

## Similar on Wunderlust

- [MLX](https://wunderlust.xyz/r/mlx.md): Stack for running local models on Macs
- [Qwen 3.5](https://wunderlust.xyz/r/qwen-3-5.md): Qwen 3.5 model family for running on your own hardware
- [Qwen3.8-Max](https://wunderlust.xyz/r/qwen3-8-max.md): Repository for Qwen3.8-Max, Alibaba Cloud's MoE model
- [Ollama](https://wunderlust.xyz/r/ollama.md): Runtime for running language models on your own machine
- [OpenViking](https://wunderlust.xyz/r/openviking.md): Context database for agents that organizes memory and skills like a file system
- [GLM-5.3 Flash](https://wunderlust.xyz/r/glm-5-3-flash.md): Z.ai multimodal model with a 1M-token context