Aider · Laras Router Docs

This guide connects Aider to Laras Router. Every step was checked against the Aider documentation.

You need an API key (lr_live_…) and a topped-up balance. New to Laras Router? Start with the Quickstart.

Set up

Install Aider

Shell

python -m pip install aider-install
aider-install

Point Aider at Laras Router

Point Aider’s OpenAI-compatible client at Laras Router.

Shell

export OPENAI_API_BASE="https://api.larasrouter.com/v1"
export OPENAI_API_KEY="lr_live_your_key_here"

Start Aider

Run Aider in your project with the openai/ prefix on the model name.

Shell

cd /path/to/your/project
aider --model openai/deepseek-flash

Add model metadata (optional)

Optional: tell Aider the model’s limits and prices by saving .aider.model.metadata.json in your home directory or repo root.

.aider.model.metadata.json

{
  "openai/deepseek-flash": {
    "max_tokens": 384000,
    "max_input_tokens": 1000000,
    "max_output_tokens": 384000,
    "input_cost_per_token": 0.0000004,
    "output_cost_per_token": 0.0000016,
    "litellm_provider": "openai",
    "mode": "chat"
  }
}

Good to know

The openai/ prefix is required. It tells Aider to send requests to the base URL in OPENAI_API_BASE.
Without the metadata file Aider warns about an unknown context window and uses defaults; the warning is safe to ignore. Aider’s cost display is an estimate; the console shows what you were billed.

Settings and screens can change between versions. If something looks different, check the Aider documentation. For other tools, see all integrations.