Aider
Use Runars as the backend for Aider's chat mode
Aider is an AI coding assistant for the terminal. It supports arbitrary OpenAI-compatible providers, so it works with Runars.
Prerequisites
- A Runars API key (Dashboard → API Keys), starting with
sk-runars- - Aider installed (
pip install aider-chat) - Credit balance on your organization
Setup
Aider reads its configuration from environment variables. Set both:
export OPENAI_API_BASE="https://api.runars.ca/v1"
export OPENAI_API_KEY="sk-runars-..."Add these to your shell profile (~/.zshrc, ~/.bashrc) so they persist across sessions.
Using it
Launch Aider against a Runars catalog model id:
aider --model openai/glm-5.2The openai/ prefix tells Aider to use the OpenAI provider (pointed at Runars via the base URL). Every request flows through Runars and is debited against your organization's credit the balance.
Troubleshooting
"Model not found": Confirm the model id (after openai/) matches an id returned by GET /v1/models. Unknown ids return 404 model_not_found rather than being silently substituted.
"401 Unauthorized": The API key is missing, wrong, or revoked. Keys are cache-invalidated within ~5s of revocation on Runars.
Aider ignores my env vars: Confirm both variables are exported in the shell session where you launch aider. Some IDEs don't inherit your shell profile—set them in the IDE's terminal configuration instead.
What's next
- Install SDKs — Anthropic & OpenAI SDK setup
- Pricing & billing — credit model, rate limits, auto-reload