⚡ Zen LM
Getting started

Installation

Install dependencies for using Zen models

Installation

Requirements

  • Python 3.10+
  • PyTorch 2.2+
  • CUDA 12.1+ (for GPU inference)

Install Transformers

pip install torch transformers accelerate

Install vLLM (Production)

pip install vllm

Install MLX (Apple Silicon)

pip install mlx mlx-lm

Install SGLang

pip install sglang[all]

GPU Memory Requirements

ModelMinimum VRAMRecommended
zen-nano2GB4GB
zen-coder-4b8GB16GB
zen-coder-flash24GB48GB
zen-max160GB320GB

zen-coder-flash uses MoE architecture with 31B total params but only 3B active, making it efficient for its capability level.

Introduction

Zen LM by Hanzo AI -- 14 frontier models for code, reasoning, vision, multimodal, embeddings, and safety

Quickstart

Get started with Zen models in minutes

On this page

InstallationRequirementsInstall TransformersInstall vLLM (Production)Install MLX (Apple Silicon)Install SGLangGPU Memory Requirements