Infer × Continue · setup guide for the IDE 2.0.0 and cn 1.5.47
Point Continue at Infer.
Add Infer as an openai provider with apiBase: https://infer.flow7.org/v1 in ~/.continue/config.yaml.
Continue is no longer maintained. Its repository is read-only after the final 2.0.0 release, and the cn CLI is frozen at 1.5.47. These steps work with those releases.
Infer Setup 0.2.0
One command.
Pick a model from the live list and enter your Infer API key when asked; cn starts on Infer with a session config that is deleted when it exits.
npx --yes https://infer.flow7.org/downloads/infer-setup-0.2.0.tgz run continueNeeds Node.js 20 or newer. Other clients and options.
Configuration
Choose the model and route.
The model entry carries the route's limits and tool support, which Agent mode requires.
Waiting for the live Infer catalog…
Put the key in ~/.continue/.env so the secret resolves on your machine: the CLI sends the name of a secret it cannot resolve to Continue's hub.
Good to know
Settings that matter.
- 01
Agent mode needs tool_use.
The generated entry lists it; images are added only for a route that accepts them.
- 02
GPT models use Responses in the IDE.
The IDE sends any GPT-5 or later model to
/responseswhatever the base URL. Infer serves it, and the entry setsuseResponsesApi: falsefor a route without it. - 03
Desktop launches miss terminal variables.
An IDE opened from the dock or Start menu does not see
INFER_API_KEYfrom your shell; the.envfile avoids that.
Start with a capped key.
Create an account, add credit, and give the API key a spend limit before the first session.