# Google Antigravity SDK adds local model support

> Developers can now run AI agents on‑device with the Antigravity SDK, keeping most token usage local and code private.

Oossa · 2026-10-14 · https://oossa.com/en/google-antigravity-sdk-adds-local-model-support

Google announced that its Antigravity SDK now supports running local models. The change lets the high‑level “architect” agent stay in the cloud while “builder” agents run on a developer’s machine. In a demo that patched three vulnerable Python modules, only about 95 of roughly 3,400 tokens were sent to the cloud, with the rest processed locally. The SDK works with Google’s Gemma 4 26B model via LiteRT and also supports OpenAI‑compatible servers like Ollama and vLLM.

## The facts

- Announcement date: 2026‑10‑14 ("Wed Oct 14 2026")
- Cloud token usage: 95 of ~3,400 tokens (≈2.8%)

## Why it matters

It lets software teams keep source code off the cloud while cutting inference costs.

## Sources & references

1. [Google Antigravity SDK Adds Local Model Support for Greater Token Efficiency and Code Privacy](https://www.infoq.com/news/2026/10/antigravity-sdk-local-models/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=AI%2C+ML+%26+Data+Engineering) – InfoQ, 2026-10-14

Last updated: 2026-10-14