Google Grants All Engineers Access to Anthropic's Claude for Software Development
Google has updated its internal policy to allow all company engineers to use Anthropic's Claude models via its internal development platform, Antigravity. While Gemini remains Google's primary foundation model, engineers are now provided per-user quotas for Claude to supplement their coding workflows. This pragmatic policy shift demonstrates the intense pressure tech giants face to boost developer productivity, even if it requires leveraging competitors' AI models. It highlights existing developer frustration with Gemini's coding capabilities relative to leading models like Anthropic's Claude. Access to Claude (specifically Opus models) was previously restricted to teams like Google DeepMind and select high-priority projects. Google emphasized that third-party models are supplied via quotas for specialized use cases, following similar internal policies adopted by Amazon.
## BACKGROUND
Antigravity is Google's agentic software development platform designed to orchestrate autonomous AI agents for complex coding tasks. Anthropic is an AI safety and research startup in which Google has previously announced plans to invest up to $4 billion. As competitive pressures mount across the tech industry, major companies are increasingly adopting multi-model internal strategies to maximize efficiency.