OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Vllm Metal

Vllm Metal is an open-source vLLM hardware plugin for running supported language models on Apple Silicon Macs. It uses MLX as its primary compute backend and provides an Apple Metal attention path, while vLLM supplies the API server and scheduling components.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Vllm Metal — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to run a supported model through vLLM on an Apple Silicon Mac and can meet the macOS requirement.

Who it is for

  • Developers building local language-model inference or serving setups on Apple Silicon.
  • Teams already using vLLM that want to explore its Apple Silicon hardware path.
  • Contributors working on Python or kernel code for the plugin.

Documented uses

  • Serve supported language models locally through vLLM on an Apple Silicon Mac.
  • Use the plugin's Metal attention implementation in a local serving stack.
  • Explore speculative decoding in an Apple Silicon vLLM configuration.
Before you choose

Requirements & limits

The source repository specifies an Apache-2.0 license. The README documents stable installation with Homebrew and a separate installer for the latest development build.

  • Requires Apple Silicon and macOS 15 or later.
  • Model support varies; consult the project's supported-model matrix.
  • The stack separates responsibilities: vLLM provides the API server and scheduling, mlx_lm supplies token-wise model layers, and Vllm Metal owns the request-aware attention path.
  • Stable and development builds have different documented installation paths.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the stable Vllm Metal release

Install the stable release on a supported Apple Silicon Mac.

Before you start

  • An Apple Silicon Mac running macOS 15 or later
  • Homebrew

Steps

  1. Add the vllm-metal Homebrew tap: brew tap vllm-project/vllm-metal https://github.com/vllm-project/vllm-metal

  2. Install the package: brew install vllm-project/vllm-metal/vllm-metal

Expected outcome

The stable Vllm Metal release is installed; the README says to run vllm without activating an environment.

Guide limitations

  • This installation procedure does not specify model selection or serving commands.
  • Supported models vary; check the project's model matrix.
Documentation 1 ↗
Source-based practical guide

Install the latest development build

Install the development build and activate its documented environment.

Before you start

  • An Apple Silicon Mac running macOS 15 or later

Steps

  1. Run the project's installer: curl -fsSL https://raw.githubusercontent.com/vllm-project/vllm-metal/main/install.sh | bash

  2. Activate the environment: source ~/.venv-vllm-metal/bin/activate

Expected outcome

The development build's environment is activated.

Guide limitations

  • The README labels this the latest development build, not the stable release.
  • The supplied setup instructions do not cover serving a model.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
vllm-project/vllm-metal · Official homepage title · retained in its source language
How we review sourcesSuggest a correction