Google adds local-model support to Antigravity SDK
InfoQ reported that Google added local-model support to the Antigravity SDK, allowing developers to run agent workflows on their machines while using cloud models for higher-level tasks. In Google's hybrid Architect-Builder pattern, the Architect plans in the cloud using filenames and task descriptions without accessing source code, while local Builders reproduce vulnerabilities, propose fixes, review patches and run regression tests. In a vendor demonstration involving three vulnerable Python modules, Google reported that 95 of roughly 3,400 tokens were used for cloud planning and 97.2% were consumed locally. A community participant questioned whether token share alone shows whether local models run faster or more cheaply than cloud models.
The reported update gives developers a way to allocate agent work between cloud planning and local execution, with control over data privacy, inference costs, network dependence and computational workloads.
谷歌在 Antigravity SDK 中引入了对本地模型的支持"，使开发者能够直接在自己的机器上构建和执行智能代理工作流，同时仍然可以灵活地将云端模型融入其中，用于处理更高层次的任务。这使得开发者能够更好地掌控数据隐私、推理成本、网络依赖以及计算工作负载的分配。
Antigravity SDK 最具吸引力的功能之一是混合代理编排。该功能将谷歌提出的“架构师-构建者”（Architect-Builder）模式应用于由云计算和本地计算组成的混合环境。
根据“架构师-构建者”模式，一个具备高上下文感知能力的主 AI 代理（即“架构师”）掌握全局视图，包括依赖关系、架构约束以及详细规范或实施计划。一旦架构师制定了计划，多个并行运作且各具专长的 AI 代理（即“构建者”）便会在隔离的环境中独立工作，执行规范中规划的、范围狭窄的离散任务。
#Google#Antigravity SDK#Gemma 4 26B A4B#LiteRT#LocalOpenAIAgentConfig#Ollama#LM Studio#vLLM#OpenAI