← Blog

Install InferAll into any project with one command

npx @inferall/cli init detects your stack, sets up your key, and points your existing OpenAI or Anthropic SDK code at InferAll — with an AI-assisted codemod for the tricky call sites.

InferAll Team

2 min read
CLIinstall wizardOpenAI SDKAnthropic SDKAI gatewaydeveloper toolscodemod

Every gateway asks you to read docs to change one line. We built a wizard that makes the change for you:

npx @inferall/cli init

That's the whole integration. Here's what it actually does — and what it deliberately doesn't.

What init does

  1. Detects your project. Node or Python, and whether you already use the openai, @anthropic-ai/sdk, or Google Gemini SDKs.
  2. Sets up your key. Writes INFERALL_API_KEY to .env, adds a placeholder to .env.example, and makes sure .env is gitignored — the hygiene step most quickstarts skip.
  3. Migrates your code. Simple client constructors are rewritten deterministically — the OpenAI SDK gets baseURL: "https://api.inferall.ai/v1", the Anthropic SDK gets the host root (it appends /v1/messages itself). Anything ambiguous gets an AI-suggested patch that you review before it's applied. The patches are generated by InferAll's own $0 NVIDIA NIM models — the wizard runs on the product it installs.
  4. Verifies with a live call. If your account isn't activated yet, it says exactly that — the $5 starter pack becomes spendable balance, not a fee — with the link, instead of failing with an opaque 402.

What it deliberately doesn't do

Using the Google Gemini SDK? That SDK speaks its own wire protocol, so no base URL can re-point it. Instead of a risky cross-protocol rewrite, the wizard prints the exact OpenAI-compat snippet (model: "gemini/gemini-2.5-flash") and leaves your code alone. Honest guidance beats broken automation.

doctor — keep it wired

npx @inferall/cli doctor

Checks key presence and validity, .env gitignore hygiene, whether every client call site actually points at the gateway, and your live account state. Exit code 0 only when everything passes — so it drops straight into CI:

- name: Check InferAll wiring
  run: npx @inferall/cli doctor
  env:
    INFERALL_API_KEY: ${{ secrets.INFERALL_API_KEY }}

models — see what you get

npx @inferall/cli models --free

The live catalog with real prices from the gateway — 120+ models, 80+ of them at $0 input/output on NVIDIA NIM within free-plan daily limits.

Zero runtime dependencies, Node 18+, and the whole thing is one file. Get a key at inferall.ai/keys and run npx @inferall/cli init in your project.

Share

Twitter / XLinkedIn

Related