Google on Wednesday globally introduced an open-source AI agent Gemini CLI for developers to help with coding, content generation, problem solving and deep research. The announcement comes close on the heels of Google Search announcing the launch of ‘AI Mode’ in India, powered by Gemini 2.5
The fully open-source Gemini CLI is integrated with Google’s AI coding assistant Gemini Code Assist, so that all developers regardless of the selected plan, receive prompt-driven, AI-first coding via both AI agents. Gemini CLI has one million token context window extending usage limits and has the largest allowance of 60 model requests per minute and 1,000 requests per day at no charge.
The agent includes built-in tools that integrate prompts with Google Search to provide users web pages and real-time, external context to the model. Gemini CLI can also extend its capabilities through built-in support, customize prompts to adjust Gemini per specific needs and workflows.
Gemini Code Assist, Google’s AI coding assistant for students and professional developers, shares the same technology with Gemini CLI. Gemini Code Assist agent mode is available at no additional cost for all plans through the Insiders channel.