Google Antigravity SDK Adds Local Model Support for Greater Token Efficiency and Code Privacy

Google's Antigravity SDK now supports running AI models locally alongside cloud models, enabling hybrid workflows that enhance data privacy and reduce cloud token usage. The SDK uses the Architect-Builder pattern to split tasks between a cloud-based architect agent and local builder agents, demonstrated with code auditing and patching. It supports Google's Gemma 4 26B model via LiteRT and OpenAI-compatible local inference servers like Ollama and LM Studio.