Add a custom model to antigravity

Prerequisites: the standalone Antigravity desktop app, Git, Node.js 22.13 or newer, and a provider API key or local server.

Install

The full sequence, step by step

From clone to first message — check first, patch second.

  1. Check

    Clone, build, then run node scripts/deploy.mjs --check. Close Antigravity and its language server first. The check verifies compatibility without changing anything.

  2. Patch

    node scripts/deploy.mjs applies the patch. The installer preserves your installation, keeps version-specific backups, and rolls back failed deployments.

  3. Add model

    Settings → Models & Usage → Custom Models → Add model. Pick a provider preset, enter your API URL and API key, then add the model by exact ID or fetch the provider catalog.

  4. Use it

    Select your model from the chat picker. Start chatting.

Commands

Run these, in order

Verbatim from the repo README. Close Antigravity and its language server before the patch step, and run the deploy commands from the repository directory.

install — bash
git clone https://github.com/vahapogut/antigravity-add-model.git
cd antigravity-add-model
npm ci --ignore-scripts
npm run build
node scripts/deploy.mjs --checkverify compatibility — no changes
node scripts/deploy.mjsapply the patch

Add your model

Where the model gets configured

The two screens behind step 03 — the Add model dialog and its provider dropdown.

Add custom model dialog with provider selection, model ID, API key, and API URL fields
Provider dropdown in the add custom model dialog listing presets including OpenAI, Anthropic, Gemini, Ollama, and Custom / Other

What do I need before starting?

The standalone Antigravity desktop app, Git, and Node.js 22.13 or newer — that Node version covers both the desktop addon and the optional gateway. You also need a provider API key or a local server to connect to.

Source: GitHub repo