LMRING

Open-source LLMs Comparison Arena

Compare multiple LLMs. Vote on responses. See the leaderboard.

English | 简体中文

Features

• Arena Mode — Compare 2-5 AI models simultaneously with real-time streaming
• Video Generation — Compare AI-generated videos across 7 providers (OpenAI Sora, Google Vevo, MiniMax, Kling, Seedance, Vidu, and more)
• Leaderboard — Ranked models with table, bar chart, and scatter plot views
• Voting System — Crowdsource model quality evaluations
• Conversation History — Save, share, and revisit your comparisons
• Flexible Auth — GitHub, Google, and Linux.do OAuth plus Email OTP via Resend
• 50+ Providers — OpenAI, Anthropic, Google, DeepSeek, Mistral, and more
• Multimodal — Text, image, and video inputs for vision-capable models
• Self-Hosted — Full data ownership on your own infrastructure
• i18n Ready — English, French, and Chinese

What's New

• Video Generation Comparison — Side-by-side comparison of AI-generated videos across multiple providers
• Web Development Comparison — Compare LLM coding capabilities for web development tasks
• Image Generation Comparison — Compare AI image generation quality across different models

Video Generation Comparison

Web Development Comparison

Image Generation Comparison

Screenshots

Arena	Leaderboard

Bar Chart	Scatter Plot	Settings

Quick Start

Prerequisites

• Node.js 24+
• pnpm 10+
• PostgreSQL database

Local Development

# Clone the repository
git clone https://github.com/llm-ring/lmring.git
cd lmring

# Install dependencies
pnpm install

# Set up environment variables
cp .env.example .env

# Run database migrations
pnpm db:migrate

# Start development server
pnpm dev

Open http://localhost:3000 to see the app.

Self-Hosted

LMRing supports self-hosted deployment with full data ownership.

Docker Compose (Recommended)

docker compose up -d

Vercel / Cloudflare

Deploy to your preferred platform. Configure environment variables as shown above.

Environment Variables

Variable	Description	Required
DATABASE_URL	PostgreSQL connection string	Yes
BETTER_AUTH_SECRET	Auth secret (min 32 chars)	Yes
ENCRYPTION_KEY	API key encryption key	Yes
GITHUB_CLIENT_ID	GitHub OAuth client ID	No
GITHUB_CLIENT_SECRET	GitHub OAuth client secret	No
GOOGLE_CLIENT_ID	Google OAuth client ID	No
GOOGLE_CLIENT_SECRET	Google OAuth client secret	No
NEXT_PUBLIC_LINUXDO_AUTH_ENABLED	Enable Linux.do OAuth	No
LINUXDO_CLIENT_ID	Linux.do OAuth client ID	No
LINUXDO_CLIENT_SECRET	Linux.do OAuth client secret	No
NEXT_PUBLIC_EMAIL_ENABLED	Enable email OTP login	No
RESEND_API_KEY	Resend API key for email OTP	No
EMAIL_FROM	Sender email address	No

Tech Stack

Layer	Technology
Framework	Next.js 16, React 19, TypeScript
Styling	Tailwind CSS 4, shadcn/ui
State	Zustand
Database	PostgreSQL, DrizzleORM
Auth	Better-Auth, Resend
AI	Vercel AI SDK
Build	Turborepo, pnpm workspaces

Supported Providers

OpenAI, Anthropic, Google, Azure, Amazon Bedrock, DeepSeek, Mistral, Groq, OpenRouter, X.ai, Together AI, Fireworks AI, Perplexity, and any OpenAI-compatible endpoint.

Contributing

Contributions are welcome! Please read our Contributing Guide before submitting a Pull Request.

License

Apache 2.0