This guide explains how to use Claude Code with Lunos so you can run coding workflows with models available in your Lunos account.
You can start with coding-focused models such as Qwen, DeepSeek, Gemma, or GPT-family models depending on availability in your workspace.
Install Claude Code first, then point it to Lunos through environment variables.
curl -fsSL https://claude.ai/install.sh | bash
Set environment variables in the same terminal session:
export ANTHROPIC_BASE_URL="https://api.lunos.tech"
export ANTHROPIC_API_KEY="your-lunos-api-key"
To persist for new terminals, add those export lines to your shell profile:
~/.zshrc for zsh~/.bashrc for bashirm https://claude.ai/install.ps1 | iex
Set variables for the current PowerShell session:
$env:ANTHROPIC_BASE_URL = "https://api.lunos.tech"
$env:ANTHROPIC_API_KEY = "your-lunos-api-key"
To persist across sessions, use setx:
setx ANTHROPIC_BASE_URL "https://api.lunos.tech"
setx ANTHROPIC_API_KEY "your-lunos-api-key"
From your project directory:
claude --model openai/gpt-4o
Example with another model ID:
claude --model deepseek/deepseek-chat
Use a model ID that exists in your Lunos model list.
ANTHROPIC_BASE_URL is set correctly.https://api.lunos.tech.claude runs.You can also use Claude Code inside your editor:
No headings found on this page.
