Infer × Aider · setup guide for 0.86.2
Run Aider on Infer.
Aider reaches Infer through its OpenAI-compatible route: the openai/ model prefix, https://infer.flow7.org/v1 as the API base, and your Infer key.
Infer Setup 0.2.0
One command.
Pick a model from the live list and enter your Infer API key when asked; Aider starts on Infer for that session. It will not start while a .env or .aider.conf.yml it reads would replace the Infer key or base URL, and names the file.
npx --yes https://infer.flow7.org/downloads/infer-setup-0.2.0.tgz run aiderNeeds Node.js 20 or newer. Other clients and options.
Configuration
Choose the model and route.
Main, editor and weak models all use the selector. The metadata file gives Aider the route's limits.
Waiting for the live Infer catalog…Aider never enforces token limits itself; the metadata file only stops its unknown-model warnings and informs its context display.
Good to know
Settings that matter.
- 01
Other settings files can replace the key.
Aider loads
.envfrom your home directory, the git root and the working directory with override on. One that sets an OpenAI key or base URL replaces the Infer session. - 02
A stray OPENAI_BASE_URL wins.
Aider's LiteLLM reads it before
--openai-api-base. - 03
Python 3.10 to 3.12.
Aider 0.86.2 does not install on Python 3.13.
Start with a capped key.
Create an account, add credit, and give the API key a spend limit before the first session.