Skip to content
项目

Choeng-Rayu/rayu-cli has 46 GitHub stars and 3 forks. GitHot measured 0 stars in 24 hours and +1 in 7 days. Last checked 2026-11-22 11:40 UTC. Last push 2026-11-21.

这些数字如何统计

实测星标增长

24小时
0
7天
+1
15天
正在收集历史
1个月
正在收集历史

各周期内的星标净变化,周期相互重叠,并非每日新增。缺失的测量保持空白。星标代表关注度,不代表软件质量。

关于

Open-source AI coding agent CLI with multi-provider support, local models, MCP, and specialized AI agents.

开发工具

README.md

原文
RAYU Code

The AI Coding Agent Ecosystem

npm CI License: MIT GitHub Sponsors Website

Website: rayucode.com · Docs: rayucode.com/docs · Changelog: rayucode.com/changelog

Rayu CLI Preview

🎥 Watch It In Action

RAYU CLI Demo

🎬 RAYU CLI in action — 4:42 demo showing multi-provider AI coding agent capabilities

📥 Download MP4 (13MB) · 📥 Download WebM (31MB) · 📖 Documentation

The ultimate standard for terminal-based AI development. Rayu is built from the ground up to outperform every other CLI agent on the market. It works flawlessly, runs faster, and respects your privacy completely.


⚡ Why Rayu Outperforms the Rest

Feature Rayu OpenCode KiloCode Claude Code Codex
Response Speed ⚡ Near-instant (<500ms) ~3-5s ~5-10s ~5-8s ~3-6s
P2P Collaboration ✅ Native Peer-to-Peer ❌ No ❌ No ❌ No ❌ No
Privacy / No Training Data ✅ Zero Data Retention ⚠️ Cloud-dependent ⚠️ Cloud-dependent ❌ Trains on your code ❌ Trains on your code
Multi-Provider BYOK ✅ Anthropic, OpenAI, DeepSeek, etc. ❌ Limited ❌ Single ❌ Anthropic only ❌ OpenAI only
Agent Swarms ✅ Built-in & Parallel ❌ No ❌ No ❌ No ❌ No
Offline Capabilities ✅ Private local model support ❌ No ❌ No ❌ No ❌ No

🚀 Key Advantages

1. Blazing Fast Speed

Rayu is built on high-performance streaming architectures, eliminating the typical latency bottlenecks of traditional CLI tools. Our terminal UI utilizes a custom React/Ink renderer with zero-GC cell buffers, and our Go gateway ensures sub-millisecond routing overhead.

  • Time-to-first-token: Under 500ms for cached sessions and 1-2s for cold boots.
  • No more waiting on sluggish web bridges or processing screens.

2. Peer-to-Peer (P2P) Collaboration

Rayu is the only CLI agent with native P2P capabilities. Collaborate directly with team members in real-time without sending intermediate payloads to external cloud storage. This is ideal for secure enterprise environments, air-gapped systems, and lag-free cooperative terminal sessions.

3. Absolute Privacy: No Training Data

While other platforms use your inputs and proprietary source files to train their models, Rayu never trains on your code.

  • BYOK (Bring Your Own Key): Connect directly to your preferred API keys.
  • Hosted Gateway: Completely transparent proxy pipelines with zero-data-retention guarantees.

4. Flawless Execution & Freedom of Choice

Rayu works perfectly right out of the box. Switch models instantly mid-session with /model or run parallel diagnostics on your codebase across multiple LLM providers (Anthropic, DeepSeek, Google Gemini, OpenAI, Kimi, and locally hosted models via Ollama/LM Studio).


💻 Quick Start

One command, any machine — no Node, npm or sudo required:

# macOS / Linux
curl -fsSL https://rayucode.com/install | bash
# Windows (PowerShell)
irm https://rayucode.com/install.ps1 | iex

Then:

rayu

Already have Node ≥ 18 and would rather use npm?

npm install -g @rayu-dev/rayu-cli

Or run instantly without installing:

npx @rayu-dev/rayu-cli

On first launch, Rayu will guide you through connecting a provider. Choose BYOK (your own API key) or a Rayu-hosted connection.


🌐 Monorepo Services

Rayu consists of four independent services. They work together or as individual standalone configurations.

Directory Stack Role
rayu/ TypeScript + Bun + Ink The CLI — AI coding agent itself
rayu-backend/ NestJS + Prisma + MySQL Accounts API — auth, plans, billing
rayu-gateway/ Go + chi + Redis AI Gateway — streaming proxy, rate limiting
rayu-web/ Next.js 15 + NextAuth Website — marketing site + user dashboard
deploy/ Docker Compose + Caddy Production — single-VPS deployment stack

🛠️ CLI Quick Commands

cd rayu && bun install && bun run dev       # Run CLI from source
cd rayu-backend && npm install && npm run start:dev   # Launch accounts backend
cd rayu-gateway && go run ./cmd/gateway      # Run gateway routing proxy
cd rayu-web && npm install && npm run dev    # Spin up web dashboard

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details on issues, code formatting, and pull requests.


🌐 rayucode.com · Issues · NPM

探索相关项目

加载中…