gemma4.devgemma4.dev
  • Models
  • Run Local
  • Deploy
  • Guides
Try Gemma 4 ↗
gemma4.devgemma4.dev

Local Runtime v2.4.0

Bring Intelligence to Source

All six Gemma 4 runtimes in one place. Install locally, no cloud required.

Recommended

Ollama

The easiest way to get up and running on macOS, Linux, and Windows. Single CLI command deployment.

ollama run gemma4:e4b
Get Started with Ollama
storage

Hugging Face

Full Python control with the Transformers library. Best for fine-tuning and ML pipelines.

View guidearrow_forward
computer

LM Studio

GUI-based local AI tool. Download and run Gemma 4 with no command line required.

View guidearrow_forward
phone_iphone

MLX

Apple Silicon optimised framework for maximum efficiency on M-series chips.

View guide →
memory

GGUF

Quantised models designed for CPU/GPU split execution across any hardware.

View guide →
code

llama.cpp

High-performance C++ backend with full quantisation control and CUDA/Metal support.

View guide →

Diagnostic Hub

Not sure which runtime?

memory

Low VRAM?

Use GGUF Q4 or Ollama's auto-quantize

computer

Prefer GUI?

LM Studio runs Gemma 4 with no CLI

phone_iphone

Apple Silicon?

MLX gives the best tokens/sec on M-chips

gemma4.devgemma4.dev

运行、部署和调试 Gemma 4 模型。专为快节奏开发者打造。

GitHubGitHubTwitterX (Twitter)Email
Models
  • Gemma 4 E2B
  • Gemma 4 E4B
  • Gemma 4 26B
  • Gemma 4 31B
  • Compare Models
Run Local
  • Ollama
  • Hugging Face
  • GGUF
  • LM Studio
  • llama.cpp
Deploy
  • vLLM
  • Gemini API
  • Vertex AI
  • Cloud Run
Guides & Help
  • Thinking Mode
  • Prompt Formatting
  • Function Calling
  • Error Fixes
© 2026 gemma4.dev All Rights Reserved.