Integrations
Google Antigravity
Use Apertis models inside the Google Antigravity IDE through the Cline or Kilo Code extension, or a terminal coding agent.
Google Antigravity is an agentic IDE built on VS Code. Its extensions come from the Open VSX registry, and it has an integrated terminal. You can use Apertis models inside Antigravity in two ways. You can install an extension that accepts an OpenAI-compatible endpoint, or you can run a terminal coding agent.
Prerequisites
- Google Antigravity IDE installed, from antigravity.google/download
- A Google account. Antigravity asks you to sign in with Google when it first launches.
- An Apertis API key, from the Apertis Dashboard
Option 1: An extension (Cline or Kilo Code)
Step 1: Install the extension
- In Antigravity, open the Extensions view:
Cmd+Shift+Xon macOS,Ctrl+Shift+Xon Windows and Linux. - Search for Cline (publisher
saoudrizwan) or Kilo Code (publisherkilocode). - Click Install.
To install from the command line instead, run Antigravity's antigravity-ide command. On macOS it is at Antigravity IDE.app/Contents/Resources/app/bin/antigravity-ide.
antigravity-ide --install-extension saoudrizwan.claude-dev # Cline
antigravity-ide --install-extension kilocode.kilo-code # Kilo CodeStep 2: Point it at Apertis
Open the extension's settings and enter these values.
| Setting | Value |
|---|---|
| API Provider | OpenAI Compatible |
| Base URL | https://api.apertis.ai/v1 |
| API Key | Your Apertis API key |
| Model ID | A model ID from the table below |
The Cline and Kilo Code pages show each extension's settings screens.
Step 3: Choose a model
Enter any model ID from the Model List. For example:
| Model | Model ID |
|---|---|
| Claude Sonnet 4.6 | claude-sonnet-4-6 |
| GPT-5.5 | gpt-5.5 |
| Gemini 3.5 Flash | gemini-3.5-flash |
Option 2: A terminal coding agent
Open Antigravity's integrated terminal (Terminal → New Terminal). Then run any coding agent that you have configured for Apertis:
The agent edits files in the workspace that is open in Antigravity.
Verify the connection
Check your key and the model before you send a prompt from the extension:
export APERTIS_API_KEY="your_apertis_api_key"
# 1. The key is valid: lists the models it can call
curl -s https://api.apertis.ai/v1/models \
-H "Authorization: Bearer $APERTIS_API_KEY" | head -c 300
# 2. The model answers
curl -s https://api.apertis.ai/v1/responses \
-H "Authorization: Bearer $APERTIS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "claude-sonnet-4-6", "input": "Reply with the word ok"}'The second command returns JSON with "status": "completed" and the reply ok. Next, send a prompt from the extension's chat panel in Antigravity, and the extension shows the model's reply.
Troubleshooting
| Symptom | Fix |
|---|---|
401 or Invalid API Key |
Copy the key again from the Dashboard. Make sure the Base URL ends in /v1. |
model not found or 404 |
Use a model ID exactly as it appears in the Model List. |
| The extension is not in search results | Antigravity searches Open VSX. Search by the extension ID, saoudrizwan.claude-dev or kilocode.kilo-code. |