gemma4-dev

xbill9/gemma4-dev

★ 0 stars Python AI/LLM Updated today

Gemma 4 on Google Cloud TPU: a monorepo of accelerator rigs (vLLM, JAX, PyTorch/torch_xla) each shipping a tpu-devops MCP server.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "gemma4-dev": {
      "command": "uvx",
      "args": [
        "gemma4-dev"
      ]
    }
  }
}

Or install with pip: pip install gemma4-dev

README Excerpt

A monorepo of **accelerator rigs** for serving [Gemma 4](https://ai.google.dev/gemma) on Google Cloud TPU, AWS Inferentia2, and NVIDIA GPUs. Each rig is a self-contained project that serves one Gemma 4 checkpoint on one hardware shape through one runtime. A serving rig ships the same shape of thing: a single-file [MCP](https://modelcontextprotocol.io)

gemma4-dev

Quick Install

README Excerpt

Related AI/LLM Servers