Introducing Support for Local AI Models in the Antigravity SDK
This is a developer-focused update — mainly relevant if your team builds AI coding agents and needs them to run offline or on a local machine for cost or data-privacy reasons, rather than always calling out to a cloud model.
What changed
Source summary
Google added support for running local AI models, such as Gemma 4 26B via LiteRT, in the Antigravity SDK, enabling offline agentic workflows and hybrid setups where a cloud model plans while a local model handles token-intensive tasks like code auditing on-device. The SDK also supports OpenAI-compatible local inference servers like Ollama and vLLM.
Why it matters to your work
Practical interpretation
This is a developer-focused update — mainly relevant if your team builds AI coding agents and needs them to run offline or on a local machine for cost or data-privacy reasons, rather than always calling out to a cloud model.
Source & availability
This is a summary of a published provider announcement, with practical context from SCALA’s feed.
Plan, region, workspace and rollout conditions may affect access in Australia. Check the original source for current availability.
developers.googleblog.comThis update is primarily for technical teams.