# vllm-distributed

**Repository Path**: mirrors_koush/vllm-distributed

## Basic Information

- **Project Name**: vllm-distributed
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-10-05
- **Last Updated**: 2026-07-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# VLLM Distributed Launcher

Run VLLM easily distributing model weights across multiple servers.

# Server 

1. Copy the `.env.server` to `.env`:
`cp .env.server .env`
2. Modify the `.env` with server's host address and vllm server parameters.
3. `docker compose up -d`


# Client(s)

1. Copy the `.env.client` to `.env`:
`cp .env.client .env`
2. Modify the `.env` with client's host address AND also the server's host addresses. These will be different.
3. `docker compose up -d`