Rayu CLI

Rayu CLI is a terminal-based AI coding agent. Bring your own API key and connect to any provider — Anthropic, OpenAI, NVIDIA, DeepSeek, Kimi/Moonshot, OpenRouter, local servers, or any OpenAI-compatible endpoint — with free model switching, full MCP support, and a complete built-in tool suite.

Educational / research use. Not affiliated with any AI provider.

Installation

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

Then start:

rayu

Update

rayu update

Or reinstall the latest directly:

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

Uninstall

rayu uninstall

Quick start

On first launch, Rayu will ask you to connect a provider and enter your API key.

rayu                      # start an interactive session
rayu "fix the bug in X"   # one-shot prompt, no interaction
rayu --help               # all CLI flags and subcommands

Inside a session:

| Command | What it does | |------------|---------------------------------------------------| | /connect | Add a provider (type → key → model) | | /model | Searchable model picker across all providers | | /help | List all slash commands |

Providers

Rayu supports four provider types:

• anthropic — Anthropic API (ANTHROPIC_API_KEY)
• openai-compatible — Any OpenAI-style endpoint (NVIDIA, DeepSeek, Kimi, OpenRouter, Google Gemini API, Ollama, LM Studio, etc.)
• bedrock — AWS Bedrock API (AWS_BEARER_TOKEN_BEDROCK or default AWS credentials)
• vertex — Google Gemini on Vertex AI (Google OAuth / Application Default Credentials)

Google Gemini is available three ways: a Gemini API key (GEMINI_API_KEY, via the OpenAI-compatible endpoint), Vertex AI with OAuth/ADC (project + region scoped), or Login with Gemini (interactive Google sign-in — free, no GCP project, via the Gemini Code Assist backend like the Gemini CLI). The Vertex/OAuth credentials also power Imagen 4 image generation and Veo 3.1 video generation. Pick the image/video model with /model_image_generation and /model_video_generation (default: NVIDIA).

Example — NVIDIA NIM headless run:

RAYU_OPENAI_COMPATIBLE=1 \
RAYU_OPENAI_BASE_URL=https://integrate.api.nvidia.com/v1 \
RAYU_OPENAI_API_KEY=nvapi-xxxxx \
rayu --print --model meta/llama-3.3-70b-instruct "summarize this repo"

Configuration

Config is stored in ~/.rayu by default. All settings are preserved across updates and uninstalls.

Documentation

| # | Document | Contents | |---|----------|----------| | 1 | Installation | Requirements, install, the rayu binary | | 2 | Quickstart | First run, first conversation | | 3 | Providers | Connecting providers, /connect, API keys | | 4 | Models | Model picker, context windows | | 5 | Configuration | Config files, environment variables | | 6 | CLI Reference | Commands, flags, interactive vs print mode | | 7 | Slash Commands | In-session commands | | 8 | MCP | Model Context Protocol server management | | 9 | Diagnostics & Privacy | Logging, telemetry, network posture | | 10 | Troubleshooting | Common errors and fixes | | 11 | Codebase Knowledge Graph | Local indexing, querying, and tracing using /graphify | | 12 | Image Generation | Built-in GenerateImage tool (NVIDIA), save/inline/terminal display | | 13 | Building binaries | Cross-platform standalone executables, versioning, install |

Issues & feedback

https://github.com/Choeng-Rayu/rayu-cli/issues