Before you start
Complete organization setup, make sure your account has inference credit, and create a key on the API Keys page. Most coding agents need a tool-capable chat model; Aider’s basic edit workflow does not require native tool calling. Image input and reasoning support depend on the model. These terminal examples use Bash or Zsh on macOS, Linux, or Windows through WSL. Install Aider using its official installation guide and open a terminal in your project. Installation instructions.Get your connection settings
Open Integrations, select a model, then choose Aider. The screen uses your organization’s region. The examples below use the public global endpoint; replace it with your regional endpoint and replaceYOUR_MODEL_ID with the full model ID shown there. Choose one of the models listed for this integration.
Set your Pearl API key in the terminal where you run Aider:
Configure Aider
The startup command applies the Pearl endpoint and key to that process. Keep openai/ before the full model ID: it selects LiteLLM’s OpenAI-compatible transport, while requests still go to Pearl.Start Aider in your project
Start using Aider
Start Aider in your project, add the files you want it to edit, and ask for a small change. Review the resulting diff. Aider uses edit formats rather than requiring native tool calls for this basic workflow.Troubleshooting
An unknown-model warning can mean Aider lacks metadata for a custom model, rather than a failed API connection. Check the exact model ID and choose an edit format the model handles well. Leave reasoning_effort unset; use the server default unless configuring Pearl’s native reasoning body explicitly.401: check the key and allow time for a newly created key to reach the inference gateway.404: copy the endpoint and exact model ID from Integrations. OpenAI base URLs end in/v1; the Anthropic SDK and Claude Code use the URL without that suffix.402: check your organization’s credit and billing status.429or Anthropic529: respect retry instructions and reduce concurrency.- For truncated output, check the output limit and reasoning budget. Increase limits only within the model’s maximum output length.