Google AI Pro and Ultra Plans Get Gemini CLI and Code Assist with Higher Limits (blog.google)

Google is rolling out higher model-request limits and broader access to Gemini developer tooling for Google AI Pro and Ultra subscribers: starting today (rolling out over 24 hours), subscribers get the open-source Gemini CLI in the terminal plus Gemini Code Assist inside VS Code and IntelliJ, with expanded capacity to run Gemini 2.5 Pro and Gemini Flash. That means more headroom for interactive coding sessions, batch code generation, and automated workflows without hitting rate caps. This matters for developers and teams using LLM-based tools because higher limits reduce throttling during iterative development, CI/CD, and large-scale testing. The update builds on May’s Gemini Code Assist and June’s open-source CLI releases and adds practical integrations—IDE mode in VSCode, a Zed editor integration, and GitHub Actions for the CLI—making it easier to embed Gemini into editor workflows and automated pipelines. Non-subscribers can opt in by upgrading to Google AI Pro or Ultra to access the increased throughput and toolset.