Introducing Support for Local AI Models in the Antigravity SDK

Google announced that its Antigravity SDK now supports local AI model workflows, with initial support for Gemma 4 26B A4B running through Google AI Edge's LiteRT runtime. The company recommends a machine with more than 24GB of VRAM or unified memory, and says the setup lets developers run agentic assistance fully offline using local GPU and RAM. Google also demonstrated an Architect-Builder pattern in which cloud model Gemini 3.8 Flash plans tasks while a local swarm of Gemma 4 26B instances executes them on-device.

Today, we’re announcing that the Antigravity SDK https://antigravity.google/product/antigravity-sdk/ supports local workflows across a wide range of local models and execution options, featuring initial support for Gemma 4 26B A4B https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/ using Google AI Edge https://developers.google.com/edge ’s LiteRT https://developers.google.com/edge/litert . The Antigravity SDK enables developers to build with the same agentic capabilities that power Google Antigravity https://antigravity.google/ . With this new support you can enable agentic assistance via local models completely offline. We’ve optimized this workflow for LiteRT and Gemma 4 26B, efficiently using the local GPU and RAM in order to further amplify what your local machine is capable of delivering Local model execution offers several advantages for agentic experiences: Here is how you can get started: We recommended a machine with 24GB VRAM or unified memory . python3 -m venv .venv source .venv/bin/activate pip install google-antigravity litert-lm litert-lm import \ --from-huggingface-repo=litert-community/gemma-4-26B-A4B-it-litert-lm \ gemma-4-26B-A4B-it-gpu.litertlm \ gemma4-26b In your directory, create a file called agy sample.py . Paste the following contents into it. python import asyncio import os from google.antigravity import Agent, LiteRTAgentConfig from google.antigravity.hooks import policy UPDATE: Point to the locally downloaded model from the previous step litert-lm import ... MODEL PATH = os.path.expanduser "~/.litert-lm/models/gemma4-26b/model.litertlm" async def main : print f"Using local LiteRT model: {MODEL PATH}. Please wait for local inference to complete. This could take several minutes." config = LiteRTAgentConfig model path=MODEL PATH .lightweight async with Agent config as agent: response = await agent.chat "What files are in the current directory?" async for token in response: print token, end="", flush=True if name == " main ": asyncio.run main In many cases we see that an Architect-Builder pattern is a great way of combining cloud model scale with local model advantages. In the hybrid demo video below, built with the updated Antigravity SDK, a cloud architect Gemini 3.8 Flash acts as the planner and conductor, while a local swarm of Gemma 4 26B instances handles the heavy lifting entirely on-device. When tasked with auditing and patching three vulnerable modules auth.py, billing.py, and database.py , the workflow maintains strict data privacy and allows us to make the most of our token utilization: Check out the example project here https://goo.gle/47cKYyV to run the built-in 3-file gauntlet or point it at your own Python modules and test suite. The Antigravity SDK with Gemma 4 26B A4B excels at building practical system utilities. In this example, the agent built a live-updating resource monitor that runs in the terminal. Given a single prompt, the agent autonomously writes a Python script that uses the psutil and rich libraries to track CPU and memory usage, generates the necessary requirements.txt file, and even tests the resulting code to ensure it works - all running entirely on your local machine and using Gemma 4 26B. cli resource monitor.py python import asyncio import os from google.antigravity import Agent, LiteRTAgentConfig from google.antigravity.hooks import policy UPDATE: Your prompt PROMPT = "Build a command-line interface tool using the psutil and rich libraries that displays a live-updating terminal dashboard. It should show CPU usage, memory consumption, and a sorted table of the top 5 most memory-intensive processes. Save the script as 'monitor.py' and create a 'requirements.txt' file. Test that it works." UPDATE: Point to the locally imported LiteRT-LM model path MODEL PATH = os.path.expanduser "~/.litert-lm/models/gemma4-26b/model.litertlm" UPDATE: Give AGY-SDK a workspace to write files WORKING DIR = os.path.expanduser "~/agy-test" os.makedirs WORKING DIR, exist ok=True os.chdir WORKING DIR async def main : print f"Using local LiteRT model: {MODEL PATH}. Please wait for local inference to complete. This could take several minutes." config = LiteRTAgentConfig model path=MODEL PATH, workspaces= WORKING DIR , policies= policy.allow all , .lightweight async with Agent config as agent: response = await agent.chat PROMPT async for token in response: print token, end="", flush=True if name == " main ": asyncio.run main The Antigravity SDK also offers seamless, plug-and-play support for any OpenAI-compatible server such as Ollama, LM Studio, or vLLM via LocalOpenAIAgentConfig . This gives you the flexibility to experiment with different local inference backends while keeping your agent orchestration, tools, and workflows completely unchanged. Get started with local AI by checking the instructions on the Antigravity Python SDK README https://github.com/google-antigravity/antigravity-sdk-python local-ai-models , and learn more about how you can run models efficiently on the edge using LiteRT https://developers.google.com/edge/litert-lm/overview . Please share your feedback and feature requests on the Antigravity Python SDK GitHub Issue Tracker. https://github.com/google-antigravity/antigravity-sdk-python/issues We look forward to seeing what you build Acknowledgements: Abhi Patel, Ander Dobo, Ben Miles, Cormac Brick, Ian Ballantyne, Jingxiao Zheng, Jonathan Reay, Kimish Patel, Lu Wang, Marissa Ikonomidis, Matthias Grundmann, Olivier Lacombe, Omar Sanseviero, Rishika Sinha, Rody Davis, Taylor Mullen, Tyler Mullen, Wai Hon Law, Xiaoming Hu, Xu Chen, Yu-hui Chen